DOCUNENT RIMUME

ED 343 586 PS 020 400

AUTHOR Hathaway, Warren E.; And Others

TITLE A Study into the Effects of Light on Children of

Elementary School-Age--A Case of Daylight Robbery.

INSTITUTION Alberta Dept. of Education, Edmonton. Planning and

Information Services.

REPORT NO ISBN-0-7732-0724-4

PUB DATE Feb 92

NOTE 68p.

PUB TYPE Reports - Descriptive (141)

EDRS PRICE MF01/PC03 Plus Postage.

DESCRIPTORS *Academic Achievement; *Attendance Patterns; Child

Development; Classroom Environment; *Dental Health;

*Elementary School Students; Foreign Countries; Grade

4; Intermediate Grades; *Light; Lighting; Nutrition;

*Physical Development; Vision

IDENTIFIERS Alberta; Canada; Fluorescent Lighting; *Ultraviolet

Light

ABSTRACTThis report describes a 2-year study of the effects

of various lighting systems on elementary school students' dental

health, attendance, growth and development, vision, and academic

achievement. The four light types used were: (1) full spectrum

fluorescent; (2) full spectrum fluorescent with ultraviolet light

supplements; (3) cool white fluorescent; and (4) high pressure sodium

vapor. A review of the literature reveals physiological and

psychological effects of light and color. Participants were upper

elementary students from five schools. Each school had one of the

four lighting types; two schools had the same type. Data on students

were collected before and after the study. Results indicated that

over a 2-year period, students who received ultraviolet 1.:1ht

supplements had fewer dental caries, better attendance, greater gains

in height and weight, and better academic performance than did

students who did not receive the supplements. Students under the high

pressure sodium vapor lighting had the slowest rates of growth in

height and academic achievement and the lowest attendance. It was

concluded that lighting systems have important nonvisual effects on

students who are exposed to them over long periods of time.

Implications for facility planning are considered and recommendations

regarding lighting for classrooms are offered. Appended are 44

references. (GU)

***********************************************************************Reproductions supplied by EDRS are the best that can be made

from the original document.

k

;c., f "1..stka-

-.7"

LISMITAIENT Of EDUCATIONDIN* et EclooNoniO Pnoonsti at* onoronentent

EDUCATIONAL RESOWICED wommartosCENTER fER101

TtengdoClocianoot Nan been reptothzed 113horn The moon of orgotnistton

onconaongD Mow ~en now boon mods to Improve

roptockucton gustily

Ponta of it** as opensonestottod In modemmint oo not noossaatdy tsprotent &OctalOERI motion cm maser

ase Of DaylightRobbery

"PERMISSION TO REPRODUCE THISMATERIAL HAS BEEN GRANTED SY

S. o

TO THE EDUCATIONAL RESOURCESINFORMATION CENTER {ERIC)."

,1 4

it-Pr lot .,,- ,4 t., . s

A Study Into the Effects of Light onChildren of Elementary School Age

A Case of Daylight Robbew

by

Warren E. Hathaway

John A. Hargreaves

Gordon W. Thompson

Dennis Novitsky

mz-t4 This study was initiated, funded and distributed by:

Cla) Policy and Planning BranchPlanning and Information &rvices Division

Alberta Education

CD 11160 - Jasper AvenueEdmonton, Alberta TSK 0L2

February, 1992

Alberta Education Cataloguing in Publication Data

Alberta. Alberta Education. Pa Cy and Planning Brawh

A study into the effects of light on children of elementary school age : a case ofdaylight robbery

ISBN 0-7732-0724-4

1. School buildings Alberta Lighting Case smdies. 2. Fluorescent lightingHealth aspects Alberta Case studies. 3. Electricity Physiological effect. L Tide.IL Hathaway, Warren E.

LB3243.A333 1992 621.322 7

Steering Committee

Dr. J. A. HargreavesDirectw, School of DentistryFaculty of Medical Sciences

The University of the West IndiesEric Williams Medical Sciences Cbmplex

Champs FlewsTrinidad and Tobago, W.L

Dr. Gordon W. ThompsonFaculty of Dendstry, Univeisity of AlbertaRm. 1043, . pharnincy Building

Edmonton, 1. T60 2N8

Mr. Dennis NovitskyRadiation Health Sennces

Alberta Occupational Ikalth and Safety4th Floor, Donsdale Place

10709 Jasper AvenueEdmonton, Alberta T5J 3N3

Mr. Roben GrantDuro-Test Canada Inc.1840, 10123 - 99 Street

Edmonton, Alberta T51 3H1

Mr. Hoang LeSchool Buildings Services Branch

Athena Education11160 J - Avenue

Edmonton, Al . T5K 0L2

Dr. Warren E. HathawayResearch and Educational Technology

Policy and Planning BranchAlberta Education

11160 - Jasper AvenueEdmonton, Alberta T5K 0L2

iii

Acknowledgements

A number of people were involved in this study and in various ways; collecting data,prepating the data for analysis, and analyzing the data. The following contributors to thisstudy are gratefully acknowledged

Information about students' nutrition was collected by nutritionists Marie Slusar andHeather Brown. Health and general development information were collected by DanaFitzrmick, a Registered Nurse. Dennis Novitsky, Radiation Health Services, AlbertaOccupational Health and Safety, performed light spectrum measuranents in all of theschools as well as performing a number of measurements on test lamps in his laboratory.Andrew Folkins, a systems analyst at the University of Alberta, provided sumort inassembling and analyzing the study's data base.

Dr. Jolm Burger, Policy Consultant, Policy and Planning Branch, Alberta Education,provided considerable assistance with respect to specifying and interpreting a number ofanalyses on the achievement and health data.

The Viur-Lite1 lamps (Le.. full-spectrum lamps which produce ultraviolet lightsu lements) used in this study were provided by Duro-Test Canada, Inc. with heado 6" at 419 Attwell Drive, Rexdale, Ontario, M9W 5W5. The initiatives of Robert Grantof the Edmonton office of Duro-Test Canada, Inc. in obtaining a supply of these lamps forthis study is very much appreciated.

The School Buildings Services Branch of Alberta Education, represented by Dr. MarianWdeschuk and Fiore Berlin, was very helpful in arranging support for the project'si"sTerent lighting systems through the Building Quality Restoration

The support provided by the principals, teachers, and students at the five study sites wasexcellent and their contribution to this study is acknowledged with gratitude. Thesuperintendents of the school districts in which the study schools were located are alsodeserving of thanks. Without agreement to permit this study to be conducted in theirdistricts, the study probably would not have been done. In order to preserve theconfidentiality of the study fiings, any mention of the names of the superintendents,principals, schools, teachers and students involved in this study have been omitted from thereport

II/ha-Lite is the registered trademark of Duro-Test Coporation, North Bergen, New Jersey.

iv

Executive Summary

Yes! Light affects people in sane surprising ways. Some wtmld even go so far as to saythat light is nutritious. In order to get a firmer grasp on some of these effects of light onpeople, a study of the effects of light on children of elementary school age was undertaken1)3, Alberta Educaticar between June 1987 and June 1989. This is the study report.

This study was intended to replicate the 1981-1985 smdy wherein it was found that traceamounts ci ultravkdet light served a) reduce the ckvelopment of dental caries. Five schoolswere involved in this study (ho speoific schools have not been named in ceder to protectthe students from undue notoriety). The shyly has been guided by a Steering Committeewith representatives from Alberta Occupational Health and Safety (Radiation HealthServices); the Faculty of Dentistry, University of Alberta; Dmo-Test Canada, Inc.; andAlberta Education (School Buildings Services and Policy and Planning Branches).

The study was designed to test the null hypothesisdifferent light types (full spectrumfluorescent, full spectrum fluorescent with ultraviolet light supplements, cool-whitefluorescent, and high pressure sodium vapor) have no differential effects on Division IIstudents' dental histories, growth and development histories, vision histories, scholasticachievement histories, or attendance histories when computd over a two-year period.

Upper elementary students were selected for the study for three reasons: upper elementarystudents were involved in the first study completed by Alberta Education, elementarystudents tend to remain in home moms most of the time which makes it easier to ensure thatthey remain under selected lighting types, and upper elementary studaus are going throughsignificant body changes (including eruption of permanent teeth).

Schools were chosen either because they had unusual lighting systmns (the case with highpressure sodium vapor site) or they were willing to have their lighting systems modifiedthrough Alberta Education's Building Quality Restoration Program aW with Vita-Lite fullspectrum lamps provided by Duro-Test Canada, Inc. at no cost to the project.

Dental ami nutrition data were collected by the Faculty of Dentistry, University of Albertaas a part of a contract with Alberta Education. Health-related data were wllected by aRegisteted Nurse under contract to Alberta Education. Attendance and achievement datawere collected and analyzed within the Policy and Planning Branch of Alberta Education.Tests of the spectral quality of the lights in schools involved in the study were completedby Radiation Health Services &Alberta Community and Occupatkmal Health.

Major findings of the study include:

over a two year pesiod students receiving ultraviolet light 4, clients developed 0.17decayed surfaces while the non-UV group developed 1.53 surfaces. Thosestudents receiving ultraviolet light supplements also demonstrated the best attendance,the gteatest gains in height and weight, and the best academic achievement.

over a two year period students under high pressure sodium vapor lightinf had theslowest rates of growth in height (2.1 cm less than students receivmg a ultraviolet lightsupplement), and achievement (a gain of 1.6 years in comparison to the 2.0 yearswhich was the average of all other sites). They also bad the lowest percentagt ofattendance (94.3 percent) while the other sites had attendance patterns of 95.9 parem,

959 t and 96.2 percent. The difference of 1.6 percent between the poorest andnext site translates into 3.2 days per year (about the thne needed to recover from a=dor cold).

These findings support the conclusion that lighting systems are not neutralthey have non-visual effects on people who are exposed to them over long periods of time.

This report concludes with a section devoted to a discussion of the ways by which thefindings of this study might be used.

Three recommendations are supported by the findings of this study. First, continued

the completed smdies should be examined with a view to providin the best 17;0144° esta

research should be carried out into the non-visual effects of light. Second, the

lighting systems for existing classrooms and new or redesigned classrooms. Third, aclearing house of information should be mated with respect to the non-visual effects oflightand perhaps calorand related implications for schools and school design.

8vi

Table of Contents

Page

Steering Committee iii

Acknowledgements iv

Executive Summary

Table of Contents vii

List of Figures ix

List of Tables

Chapter 1 Purpose of Study 1

Chapter 2 Literature Review 3-Related Studies 5

Al s LiOtiColor Study (1981 - 1985) 8Findings 10

Summary 11

Chapter 3 Study Design 13Study Sites 13Study Population 14Data Collection 16Corerolling for Bias and Contaminating Variables 16Dental Histories 17General Health and Growth and Development Histories 17Vision Iftstories 18Anendance Histories 18Scholastic Achievenmit Histories 18Nutrition Histories 18Skin Types 18Data Processing 19

Chapter 4 Study Findings 21A Description of the Study Population 21

Age 21Sex 21Nutrition 23Fluorine Levels 23

Study Findings 23Dead Effects 28Attendance Effects 28General Health and Growth and Development Effects 28

Age at Onset of Menarche 28Height Gains 30Weight Gains 30

vii

Page

Body Fat Gains 30Scholastic Achievement Effects 30

Effects 33Vision Wm's, 33

Fluorescent Lamps as Ultraviolet Sources 38Risk of Exposure to Supplemental Uluaviolet Light in Classrooms 38Sumueuy 44

Chapter S Conclusion and Recommendations 47Conclusions 47

What Do We Know About Light and In Effects on Students 47What Don't We Know About L' _i t and It's Effects on Students 47Putting the Study Findings Into - e 47

Recommendations 51

Bibliography 55

1 0

List of Figures Page

Figure 4.1. Changes in uncorrected distance vision (both eyes) asmeasured by a standaid 20-Foot Snellen Eye MM. 37

ill

List of Tables Page

Table 2.1

Table 22

Table 3.1

Table 4.1

Table 4.2

Table 4.3

Table 4.4

Table 4.5

Table 4.6

Table 4.7

Table 4.8

Table 4.9

Table 4.10

Table 4.11

Table 4.12

Table 4.13

Table 4.14

Biologic and medical effects of light. 9

Reflectance of various matetials for energy of wavelengthsin the region of 253.7 nanometers 9

Measured light in classrooms. 15

A two-tailed t test comparison of the age of studyparticipants as of JIM 30, 1989 22

Sex of stuly participants. 22

A two-tailed t test canparison of daily nutrition levelscalorie, protein and sugar intake. 24

A two-tailed t test companson of daily nuuition levelscalorie, protein and sugar intakeft UV enhanced sites andnon-U'V enhanced sites. 25

Effects of supplemental ultraviolet light in classrooms ondevelopment of dental caries in elementary students inGrades 4, 5 and 6. 26

Summarized effects of supplemental ultraviolet light inclassrooms on development d dental caries in elementarystudents in Grades 4, 5 and 6. 27

A two-tailed twst comparison of rates of attendance 29

Incidence of girls reaching the onwt of menarche 29

A two-tailed t test comparison of height gains measured incentimeters between June 1987 and June 1989. 31

A two-tailed t test comparison of weight gains measured inkilograms between June 1987 and June 1989 31

A two-tailed t test amparison of gains in s fat measuredin millimeters between June 1987 and June 1 6 9 32

A two-tailed t test comparison of total achievement gainsmeasured in grades between June 1987 and June 1989 34

A two-tailed t test comparison of skin types in relation to theincidence of dental caries. 35

A two-tailed t test comparison of pretest vs posnest changesin uncorrected visionboth eyes tvgether. 36

Table 4.15

Table 4.16

Table 4.17

Table 4.18

Table 4.19

Page

A pre-test/post-test comparison of the percentage change inthe number of students never using glasses. 39

Measwed radiant energy from a sample of differentfluorescent lamps

A comparison of new lamps and lamps that have been in usefor fifty (50) months at study sites 41

Relative safety of IN supplements in classrooms. 43

A summary of significant findings of the suxly together witha description of lighting system characteristics. 45

Chs[pisff ii

Purpose of Study

Education is a multtfaceted process which includes opportunity,environment, curriculum, teaching dectiveness, parental support, innateability, and physical factors. If deficiewies exist in any o f these areas, achild's education may be COrnpnwised (lielwrston, et al, 198 5: 346).

This study was undertaken in response to two issues: (a) the view that envitonment is animponant facet of the multi-faceted educational process (Helveston, et al, cited above), and(b) the findings of earlier research carried out in Alberta. The most surprisinkfindinfrwr the earlier research supported by Alberta Education from 1981 to 1985 (1986; Hargreaves & Thompson, 1989) were that those elementary students (Grades 5 and6) who received trace amounts of ultraviolet light in their classmoms developed fewerdental caries and had better attendance than students in a cm:poison mug.

Based on the 1981 to 1985 research it was comluded that then could be significant benefitsfor educatice in several areas. On the basis of the daily per pupil educational expenditure in1984-85 (i.e., $21.42 per pupil space per day) and a With difference of 949 days ofabsence per year for students under different lighting systems, the cost of having thesespaces vacant because of these absences amounts to $203.28 per pupil per year (i.e., anaverage expenditure of $20318 is made to provide a pupil space which is not used). Onthe basis of the reduced dental caries, a further saving was calculated at $116,75 per pupilper year. Combined, these benefits total $320.03 per pupil per year. Inasmuch as the costof providing trace amounts of ultraviolet in classrooms amounts to no more than $30.00

arrdent per year, there is a net social benefit of $290.03 per pupil per year. It wasconcluded that if these benefits could be generalized to all students in Alberta

(approximately 430,000) the net social benefit would approach $125,000,000 per year.Moreover, these findings may be generalizable to other regions of Canada, the northernUnited States, areas frequently shrouded in fog, and other geographical locations where thedaily availability of natural ultraviolet light is minimal for extended periods of time.

Even though the findings of the aiginal studies by Wohlfarth (1986) and Hargreaves andThompson (1989) were highly significant the bottom line was that further research wasneeded in mkr to confirm the original findings.

At the same time the literature on the non-visual effects of light on people is constantlyexpanding. Research at the Massachusetts Institute of Technology (Ozaki & Wurtman,1979: 339-341) drew attention to the fact that light from high ressure sodium vapor lampsproduced anomalies in the growth and development of animals. In addition to highpressure sodium vapor lamps, schools in Alberta and elsewhere are dy lit with avariety of other lamp types: incandescent lamps; cool-white, warm-wl; full spectrum,and UV-enhanced fluorescent lamps; and mercury vapor lamps.

This study was undertaken in order to (a) replicate the findings of the initial studyundertaken by Alberta Education into the effects of classroom li :li ting and (b) to test thedifferential effects of other lighting types on students. It was p , around a pretest-

posttest, multiple-group design and attempted to test the following null hypothesis:

4

Different light types (full spectrum fluorescent, full specaum fluorescentwith UV enhancement, cool-white fluorescent, and high pressure sodiumvapor) have no differential effects on Division II students' dental histwies,growth and development histories, vision histories, scbgastic achievementhistories, or attendance histories when compared over a two-year period.

1 52

Chegraw

Literature Review

The environments in which we spend significant parts of each day are a far cry from thenatural environments of our ancestors. We fmd ourselves sunutuuled by walls, floors andceilinv; covered with colors sekiom repeated on the same scale in flat= and these colorsare usually perceived under ting systems designed more for efficiency thmi for theirpossible physiological or ps logical effects at people. Indeed, our artificial lightingsystems can only simulate twilight levels ci illumination (i.e., light levels of 200 to 1500lux in comparison to natural light at twilight of 2,800 to 8,200 lux and at noon up to100,000 lux (Thorington, 1985: 44)).

Corth (1984) might argue with some of these views. His contention is that the naturalenvironment of our earliest ancestors was the forest floorand not the open plains. As aconsequence the habitat noon-time light levels would have been much lower than the 8,200to 100,000 lux found in open areas. Moreover, he contends tha the spectral quality of thelight at the forest floor was the combined result of the solar mdiation spectrum and thefiltering effect of the forest canopy. He says:

Field measurements of the spectral distribution of the light under a numberof forest canopies have been mode by Federer and Tanner. Although tiwreis some variation with species, all show a similar structure since chlorophyllis Me primary component in the foliage which determines the spectralabsmption. Regardless of the variation in the spectral characteristics of thefoliage and in the density of the forest canopy, the light that comprises theforest illumination is markedly different from that which enters the top ofthe canopy. A more dense canopy would skapen the curve around themaximum near 550 nm and a more sparse canopy would flatten it.Nevertheless, all reasonable proportions of transmitted, reflected, andunaltered daylight lead to similar results, namely, a 'greenish-yellow' lightwith a maximum, in the visible, near the peak of the photopic spectralsensitive function of the diurnal primate, including man.

Corth (1984) further argues that our ancestors originally occupied the forest floor near theequator. Only later did they move onto the open plains either north or south of the equatm.Thus he concludes that heavy skin pigmentation was a matter of camouflage for survivalmore than it was a filter against ultraviolet light. As humans left the forest cover andmoved into the more open country to the north (or south) and away from zones of intenseultraviolet light (and into zones where body covering wa; necessary to ward off the cold),the pigmentation ritcreared as a re se to the need fur incrmsed vitamin D which isformed by the action cf tatraviolet liton the skin.

Following Corth's logic, one might expect two effects. First, inasmuch as the lightof cool-white flutrescent lamps approximates that of the light reaching tat forest

oor, people may find these lights to be very comfortable. Second, if sldn pigmentationdecreases as a response to an increased need for ultraviolet light, highly pigmented peopleliving in northern climates may have girater needs for ultraviolet Wmubdon (or vitamin Dsupplements) than do lightly pigmented people.

16

For people, sunlight is by far the most sit. t some of light and energy and it may beexperienced as direct light or as skylight diffuzd light). Most people spend part of eachday unckr the influence of sunlight However, as society becomes mon urbanized, peoplespend much less time under sunlight and nuich more time under artificial lamps.

Although there has not been a great &al of research into the effects of light cm students, areview of some of the more basic research that has been conducted provides a perspectivefor considering the human element in school facility design and construction.Consequently, the following discussions will center on natural light, incandescent light,fluorescent light, and other sources including mercury vapor, metal halide, and highpressure sodium vapor. It is also important to distinguish between discussions aboutlighting sources and discussions about well-designed lighting systems. As the latter is awell-documented science, it is not discussed here. Rather, attention is given to somees andqualities of diffmnt light types.

Danzig, Lazarev, and Sokolov, cited by Hughes (1981: 5), captured the essence of muchresearch into the non-visual effects of light They contend that physiological disorders mayoccur in the human system if the human skin does not receive some exposure to solarradiation, either direct or diffused, for long periods of time. They believe there will be avitamin D deficiency followed by weakened body defenses and an awavation of chronicdiseases. Wurtman and Neer (1970) suggest that non-visual rednal responses to lightmediate a number of neuroendocrine hormonal functions which in turn regulate suchmechanisms as pubescence, ovulation and a wide variety of daily rhythms. At about thissame time (i.e., in the early '70s), a national committee of physicists, chemists andphysicians in the United States ma& it clear to the medical community that "photons mustbe considered a drug" (Lamola, 1985: 122). Faber Birren has been quoted as saying thatultraviolet radiation intaisifies the enzymatk processes &metabolisms, increases hormonesystem activity, and impmves the tone of the central nervous and muscular systems (Roily,Bursar, & Musemeche, 1979: 16). More recently Downing (1988) argues:

Them is no area of our mental and bodily functioning that the sun does notinfluence. Ow bodies were designed to receive and use it in a wide range ofways. We were not designed to hide from it In houses, offices, factoriesand schools. Sunshine, reaching us through our eyes and our skin,exercises a subtle control over usfrom birth to death, from head w

There is wide consensus that sunlight nurtures most living things and some light-relatedprocesses are well understood. Photosynthesis (the ability of plants to use sunlight inmanufacturing food), phototropimn (the tendency of plants to grow towards light), and theeffects of light on vision, are examples. The non-visual effects of light on people are notso well understood.

Since sunlight contains all colors in relatively uniform amounts (i.e., there are neithersharps peaks nar discontinuities in the spectral disttibution), all colors arc equally visiblewhen illuminated by sunlight For this reason natural light serves as the reference forcomparing the color rendition characteristics of other light sources, with natural lighthaving the maximum or refezence Color Rendition Index (4IMD of 100. The color renditionindex is a measure of the way colors look under specific light sources. It is important tonote that equivalent CRI indices main the same thing only when the light sources to whichthey relate have equivalent color temperatures. As a consequence, colored objects may

different when viewed under lights with different color temperatures but equivalentindices.

Not all artificial light sources accurately 61 11 ce the full specmim of sunlight.Incandescent lights am rich in red and yellow 2,1 but radiate relatively little energy in theblue and green of the spectrum. Cool-white flumescent lights emit most of theirradiant energy in E. green and hands of ills spectrum and eyes are most semtitive tolight in this range. Indftd, as (1985: 50-51) points out, it is at to 555 nut (theyellow-grew range) that the lumen or the standard unit of light is defined. MI spectrumlights emit a significant portiw of their radiant energy in the blue area of the spectrum Afurther small percentage of the radiant energy from fluorescon lamps may fall into theultraviolet range. Since the eye is less sensitive to blue light than to green and yellow light,rooms lit with full tt fluorescent lights am often peiceived as being dimly lit Fullspectrum lights do, - 9 have a relatively high Color Rendition Index and this may kvery important to vision processes.

In this regard, Aston and Bellchambers (1969) compared high efficiency lamps (i.e., I I,

such as cool-white fluorescent bmps) with lamps that provided a spectrum more 4^ ybalanced to natural light. In their report they say, "The results clearly show that the9Colorite lamps [lamps simulating natural light in spectral distribution] not only providebetter colour qualities but give a higher degree of visual clarity than do the high efficacylamps at an equivalent illumination [added by author].

Light-Related Studies

Aside from enabling vision, there are some interesting non-visual effects of light onanimals and people; some of these effecn are physiological and some may bepsychological

By 1919 researchers had reached the conclusion that sunlight was the key to the cure ofrickets. It was later discovered that artificial light was also effective in curing rickets(Loomis, 1970: 77-91). Today it is accepted that ultraviolet radiation derived fromin the regiw of 290-315 rim triggers the devel -4.1 of vitamin D in the skin (Neer, 19 5:15; Holick, 1985: 4) and vitamin D formed in I skin or regular doses of vitamin D takenorally can prevent or cure rickets. Other studies point to different effects of light onanimals and people.

Zamkova and Krivitskaya (1966: 41-44) augmented regular fluorescent light withultraviolet suntan lamps in a controlled experiment involving school children. Theyreported interesting finrlings. When compared to the control group, students who receivedexposure to ultraviolet light showed increased levels of working ability and resistance tofangue, improved academic performance, improved stability of clear vision, and increasedweight mid growth.

Volkova (1967: 109-111) studied the effects of ultraviolet supplements to general lightingin a factory. Compared to a control group, an experimental group of adults demonstrateddecreased permeability of skin capillaries, increased white cell activity, and reducedcatarrhal infections and colds.

Richard Wurtman (1968) concluded that light has biological effects that are important tohealth and that some of these effects couM be easily reproduced and measured in theexperimental laboratory. These effects were of two kinds: those which modify theindividual's endocrine, hormone and metabolic state by means of light reaching the retinaand those which result from lig,ht on the skin (e.g., vitamin D production, skin tanning,and dissociation of bilirubin). Wurtman (1969: 32-37) also linked light entering the eyewith responses of the pineal gland and secretion of the hormone melatonin. This hormone

158

in turn influences the functions of other glands, possibly as a result of direct action onspecific areas of the brain.

Wuriman and Weisel (1969: 1218-1221) studied the effects of light from cool white lampsand full-specmun Vita-Litet lams on a gmup of rats. Their findings 41,. the argumentthat environmental lighting has an effect cor at least some mendocrine

Hinsnelfarb, Scott and Thayer (1970: 1218-1221)spectrum) lamps was sivificantly more effective incod-white lamps. They say:

In conclusion, it may be stated that Vita-Lite lamps used at intensitiesrecommended for ilhanination produced signfficant killing of S. Aureusover protracted periods of exposure. High levels of such artificial daylight,including the small amount of UV normally present, could thus be of vahtein controlling air contamination with this and presumably other organimswhere danger to personnel precludes the use of standard bactericidal lamps.

tlitt+0,10iti,

that light from Vita-Lite (full-tucteria than light fiom standard

Downing (1988: 84) offers evidence that small =vaunts of UV radiation datroys bacteriaand moulds. At an irradiance level 10 ttW/cm2 there is sufficient enera to kill in oneminute a variety of bacteria and mouldsalpha strep. and stapk aureus (100%), betahemolytic strep. and E. coil (85%), moulds (30-65%), and the AIDS virus (20%).

Even though there appears to be some capability for the UV component of full spectrumlamps to have a germicidal effect, the greatest effect occurs in the UVC bandespecially at2700 nanorneters (Bickford, 1981: 19-14). Emissions at that wavelength are about 100times mate effective than emissions in the UVA band.

East (1939) reported a correlation between annual hours of sunlight and the incidence ofdental caries in 94,337 boys (ages 12 - 14) in 24 American states. Sharon, Feller, andBurney (1971: 1427-1431) found a relationship between artificial UV and development ofdental caries. They reported that golden hamsters exposed to Vita-Lite fluorescent lamps(with a measurable UV output) had one-fifth as many tooth caries as animals exposed toconventional fluorescent lights. As well, gonad, submandibular gland, and total bodyweights were greater for animals raised under the simulated natural light (Le., full spectrumwith measurable UV).

Neer and others (1971) described a study which involved elderly male residents at theChelsea Massachusetts Soldiers' Home who were exposed to full-spectrum fluorescentlighting. These lights delivered about five percent of their total radiant energy in theultraviolet range from 290 to 380 mn. The studys =elusion was that relatively smallamounts of ultraviolet light can stimulate calcium ab among elderly men who haveno exposure to sunlight and who eat a diet containing e vitamin D. It is important tonote that during the course of the study the experimental " was exposed to 500 foot-candles (5,500 lux) of illumination from Vita-Lites for ei t hours a day while the contmlgroup was exposed to 30-50 foot-candles (330-540 lux) of illumination from cool-whitelamps.

Bilirubin is formed by the breakdown of hemoglobin and is normally excreted afterundergoing a chemical change in the liver. Many 11! vborns suffer jaundice from a

1 Vita-Lite is the registered trademark of Duro-Test Cognation, Ncrth Bergen, New Jersey.

6 1 9

I lit. ... . 1* L414.1, 14 9 I ' 121.4 1 81 11 .0/4141 . I 11111 '1191 9 81 .../1 I I 4 1 ;II It 4 1 40 8 .94

4 1 4 / i t 6 1 4 . I I 9 1 I I I I , 9 0 1 4 1 o, 4 1 11%41 1 li I I I I 4 1 1 1 4 1 1 4 1 . " 1 1 1 ' 1 1 _ . 1 9 / 1 1 ' .4 1 I I I b J49 111 4.... I .1 II 94 44 .. If 1 II 1 I I 94 01 411 4 t : illz.,±al ." 141 11 16..4A I I .1 1 II 4t.41 ..1 fit41 11 I a 4 18 s1 I t 4' 19 It 0 1 I 1 t... 4 8 I 4 I I

II 1 ... 1.1 11 ... P.,Z, AL 91 it ...9) I 4, I .' e 9 't 1 11 It II 9 1 .+,11 II .411 1 ..+I1 1/I :4 M.11 1le 9,_i_c±.1.,-+; 1' t _LILL_ 1.1 V a ..i

9: 5411 . 1,1 I 4

i het 4141 1 I I 1.1 I 4., I '6 1 : I 1 I - VI 10141I,4,1 1,4t1:1 1 / I 4.- I I ' 1 4 fAIL8111 8.211....1111

Ill 1 *II. 1.1 1 . 1 0 ...a 6.1; : I :.:( i ;01 t..:1,4)/ ', tg "-Ili 1kt 'le 44 I li i :41 1 1 1 1/, ' /l'a ' I 1 II t(611,41 i 4 I ". I I ' 4 4,4 1 04; 44 Vt. .1 t_Li a : It) 41I It/: I

* 1 I 9 4 : 5a

a e :I:Al:V.42 1 I ' I ' 6.4.411i11.11 I 1 '41t 4...11± t 9.4i .4901_

'..1 1 cii -tilt it; 1 9 I b.k. 1 .,..1 %JO Lim' 1 .4/ .01 A 88 I ...,AIII , I.-8 I II I 1 1 :

1 0,9.41199)8 III 1 414......Mit 1st : Ult.,. I ,r, II 110 4 et 6 II 1.., I 1 111949" li 1194114 ..,4 4911 II

.41 tool I 1 1 :I ' 191 1 bk. 9 1 1 4 If ad ' I 1 nil I II 'wt. AP .... /...1.0. 1 1 %(...k. li1 1 11 ; 44 11

4 11t 1 .4 41A4)119.9. 1 4 46 I I 1 9 . I ' 1 4 6.: 4

6....10111 II ' IA 44 4 I I ' 1 0 1 4 4 4 1 4. I 4. : 11, I ; .4 -I I, 11111 I

1.41 IP: I9._.J 1 1 .11 9.111 A 1.4 1 ' 1110 ...1 8,..,1119A. 111 :9 1...' I ..4 If '111 . 1.9.. *0.41 .. 6 4.1

6 111. I 6.1 144 : ..4 54 4.$ ($ 1 4 4 41 I 6 It 14 III 1 141 "1 1 III . 0 4 6: dr 1 .I r ...i.... 88 4 I 4.1 ' 54 ' 1 f 144 : I 4 14_4;1 11, 11 I ',. 1 1 144 4 0 0 1 I

1 I 1 I 1 ' 1 0 1 44 4 4 1 11t.A4.4 :.1_1,'_,41 It 1 A 1 1 1 I 4..: I 4 I ' I 1.4 .0,1 441141 lie '.111 I.

6 00 . ' I 441 . / 1 .4. 4 6 I 4)_ 14,4 :t, , ' I 4 1' let 0

al 1 1 ..t 1 1. 4 1 0 4 II : I 6 I II 1 : 4 I I I4 I

I . I 0 1 $ 141 _ 0 ' 0 :4 I 11. 11 4 8 4. 44 1 1 I

: 1 4 Pk 11 I 0 : 6 0 1: 11 I I 11/

1 I 1 I 1 0 I I 4 4 V 1 1 I II IL...i_L it o $ ... 1 448 41 4 4 11 . .9 :F

1 4 '444 1 I 1 4 , 1 4 4 ' I r 1 0 1 / , 5 . 1 I 1 I V 4 1 V .. I . ...t, II W.:: I 4 v. ....

040111 I 1 1 41 Iii I 1 V 1'1 I 1 4 4 f: 1 1 ; 11- . ii 6 4 IA 4:

Ai 4 8. 1111 1 II 1 1 46 I I' IP 1 : NIII 4 8 4 , :11 441111 t' 1 1.1

8 .I . 8 sii t ;.. i . 1 1 A 1 . 4 I 1 6 1 40 I 4

1 4: 1 . 1 . `0. 1 . 1 1 O . 4 . : 5 1 " . 1 4 I _ 1 : 12- 1 1 1 I I I I ' 1 6 4. .1 8

I 411 I O I1) II 4 4 4. 1 4 4 I I 4 4 6 kt, 4 I 1 6 1 4. 1 V ; :1 I:°.:/.1. 1 4 I Is.. .0: .., 1 I . 1 ILLIII 4 /

Pi I ,..

4 :lb, 4, i 1 4,41t,t.t6 4 . '01 III I 4 4 11 4, 1 I*"

:

* II a.5 5 0 .s : s /tie : I )1. 1 1... II 41$ 0 1

4 4 I 4 O 4 4 4 I : 1 )(...., 0 4 I 1 . 6 0 I 1 44

1 '1 . 141 11 1":13:. tIAL/1 1 : 11 4 11 1.1 1 1,411 I

11, I I I 4, 5.4 I I I I 4 1 .4 t . 81 I I II

I 1--1 I.. I . , 1 I I It I I II I I I

I ' I I I I a I., I f f t 2 1 .1 ,A. II I I I ' 2 1.11,10 , s I

I I . r , II I . I 1 I I I I

III p * I 8 '80 I t 0 4' I I -IP

I I 8 I I I 1 I i

4: 4 11,19 4 I ' 4: .8 51.... 10.4 I 4.

I to -.It in : I I MC; 5 :

1 bill I 011 5 001" 1

To 'a .- Sr 14 47.--A' 4- a .- a 4 ' a 1 7-1-1,-, - 4-"''-:-. 1-1:4712 0

_ ,

1 a - , 11 ' 1 11 TT" 1 : 1 1

. 4 -1-k-v a a 1T4 111 6 I

- --.iic, i a . 1.W...ar, 1 1/11 a,.; lri , a al r.", a : : 4 1 '

a7-a a" a to

I ,

; a a

a 0: 1 : 1 la it 004 ; 14 1 ; . , .. 1 I r .11 : 1

11 1 1 g a a 1 1 14 F. 1 1 11 4 11 -.14.1,11 I 1 . i IV fr 4 1

t

, a a 4 13 I I 1r4" 1 i ,:' -1-3T 41 - .4 / 4 A 1 I .- F. , " ii,-,---.

a ,,,iaro.- ' 1 . a ta a -41, a 4 ... s a a ...... a

a . 1 }-, 1 ;

IC I Ill 11 ,-1 ' 1 1'7.

1 al ' 1 ,..4.- 1 1111-: .."4^ /

Oran- I- ,

Tro a i a;-;

1-.1 ' a la o t 1.-

... a a a a i o: al 1 II144 1 aT-v- 4 rif ,.. I : . .

6 .

1 .411

1 i a 11 :la 4 1 ,a a 1 1111 1 1 4." :-.-4 1" , irt - 1 1 0 1

46a ;.: a 4 a a a 14+4 1 11.41 4 1 11 A II ; , , . , in .... . ,,, 1e 11 a 4 t t

I r ,r-,,,,Tiri 4 41 4 44 a a 1 11 4 1 1 44 1 - I .: 44

'17-4:-41-1:4- : all all:

-; 1 ,i 1 " 1 4 1 ,

1 : 1 1 .

11; 1

4 11 4 a - I i7 ca, a 4111 ' -V.- 11 t I 11 4' 4111"1 - 14 44 111 *-1717111747 4 .1 4" ,

'ea 1 1 -.

.11, 14 1 I a 4 r , .----7:-,4 IT-7 a . o

, ogb iri : , 4 V 1 4 1 .,.

-lot il 1, ..:: : , 1 a 4

7 I -II la-J., . to t ,

. , 1.%-iwt1' II 1., 1,- 1 .-f la 4

a ,, I 1 a 4 %. - t 11 I I , , 6'1 1 4 C a 1 4 1..," 1 -11-41L 1 1 fa

1 111 ,111 III II: , 4 7 _a a="'" la 141.1- .4 114 1111 (1,17117717-

. a a ,

a 11

4 4 1 ' 14 4.4 1 j 11" 4. -, 1 '4 1 : ,1 a 1 1 1 11

, ! 4 r: a f 763 r4 ha.'a a

:6 Ca 4-.4 I ,1 a at a 411, -, 1 a 1 e a 04 1,: 4 I 11 1 4 a 1,11.4 1, 1.,

1 1 8-7.---;'4 a ; a a at 1.4 1.1111c. 4 11 a 1 a ) 4.4a I t 07- 141 a 1 V.

I ,..441 a a ,

1 1

1-4"4 b., 4.1; ,4 I a-a ip 'a at 41 I ; 11

- ,4$6 I 4 1 a . f a I .

ar- .a

41 ",,: 1

1 4 ' a '';`".4'17"--,N4 : 4 at: ':^ 14 :4111^11 II -%47-t" 1'14 1t 1 4 lp-, 11 1 1 1; 11)" 11 1 6;

_ 4 r I IC.' -,..-

, 6 ; .1 1 '-, 4 4 / 1,4 , I 1 11 : 114.1-111-11 1 -. 11 1

... .., ra a W . . . art a a . t a a

, 1 4 1 ; 1 4 .:".." 1 1 1 1

, 1 7 ! 1 . - 4 1 - C - r ; 1 4 I : S' 1 r 1 1 4 : 1,

11 1."`e.S.-4-r-1 4 1 ' 4 -fart .0 a "ilakT 1 a : 1 a 1 111 4 . 1 '44 4.110 al, a ' To , ri a 1

1 1 4 1 0 1 7 a a ,-- 1--e-o-a a a , a a ,

o .

a a a l VI ' I 1 7 i t e a 4 ; 1 1 I a 4

.V 1-4 01 14111 -41114.--' ,

1(4111 1 1 111 `I " 74 4 tal a .

1'11 I -

I 41 I 1; ,4 1 la -11' 1,11 a 4 a ,

I 4 II 44 I -:

(4 IA a a : ' /a 14 .-) - t 1 a 1 .

4

.. '1 W- ITC- t 1.-"1 111 - 't- I a ' 1 ,, a a. .4 .. a 1111 : 44i 7 a a 1.7.-a a a ,

a 1 4 4 4 1 44

: -.7 a a to 4 I : CT' II I a a i a ; f I 4 4 ; as la a

, a Tirl(6174 a I a .1 a a i 17-1 a

'pill.. t.IN le- 41 ta V.-1-1;.t a Veil s-va a I :. I t a 1 .1 a W .

;t; a lir. 1-ra a t. .: ',7t 1 17..1 ta` -1 a

4 I

1 r 1'4 1'1 : 1.1"."T",V4'.1 1 447-'111 '1 " 1 14 74- (it' .1 t' vat . - . a . .

67-114 4 1)1(11-li ,

14 1-74-C-) 'I 1 a I a I Wi . 4-7-11 .

a i-1-iii a -. a a t a7111 o a Wro a

a

4 ; - '11141 ;al 1t4 ,: 1 a era ar: . r7 o ,

a , oil ta vt motto,. 1

.-- - ar7 I "1 a i ta ' r. a a 4 111 : -4 II.- Is 1) 1-1 4-1 4 ii , , 114, I :411..14 4 14.- a 411 a

. a 4 OA .4 11t 4 ,.. /a s ,44 44- .4 -1-/ rt .-.1* I . MU 1

.1. I 4

4 'in, : 14 : 4 i s , I I I I till : i . ri", $ i, $

iri-' r`T r 1 a. 1 I 1 I / I I * .1-7-11. I

114 .o1 TF--71; A A 41 ,1, , I riF 1 r r r

r #7 I, ITI'.4"4,4,- I as... ,,, I I, 4 i t 1,

a a.

a I 47- , a , -41=ti a 1 I

r1 iro ' 5 1

t To a, a, ,1 14 , a

ii : 1 a a 1 1(1 1411 s 1 la , 11f 0-

1 1/ 1 -4 1 4

Table 2.1. Biologic awl medical effects of light (fcorn Wurtman, 1985: x).

Direct Indirect

Physiologic

Pathdogic

nowt&

ErythemaPigmentationEpidermal thidceningVitamin D synthesisBlood levels of amino acidshnmune systems

Photosensitization porphyrinsdrugs; toxin' s

Ocular damageCarcinogenesis

HyperbilimbineiniaRicketsLight-dmg interactions (Psoriasis.Leukemia)

VisionEntrainment of rhythmsReproducdve activationEntrainment and suppression of

melatonin synthesis

(Poorly characterized behavioraland medkal effects)

Dewession: seasonal affectivedisorder (SAD)

Jet lag

Table 2.2. Reflectance of various materials for energ of wavelengths in the region of253.7 nanometers (from Bickford, 1981: 19-17).

Material Reflectance (percent)

AluminumUntreated surface 40-60Tamed surface 60-89Sputtered on glass 75-85Paints 55-75

Stainless steel 25-30Tin plate 25-30Magnesium oxide 75-88Calcium carbonate 70-80New plasm 55-60Whim baked enamels 5-10White oil paints 5-10White *star paints 10-35Zinc oxide paints 4-5

9

22

.11, 6-4 4, It t VII. 1111 I '41 i.tioo 4111 - I Iit 4 tvii to 4,1,10 ,' I 1,..4114, I It 4_-.1 I I 14 ' .4

II.:,, ..... '1.14 3 4.5 I I 1..0 11 14 11/, . 1 : 1,1 _AAP _f ,: 1 02._ 1 1 . ...I.k.),..

'._ 1111_2 / fl. 4 1 I 4, * .. ,* '. _. .... 6 I 4 0 5 .4 :Aft ._12,4 II 4,- (.-2_90 o I ii ti,1 I

II I *I -1+4,-,'

' 1..6.... $ .... II I I ,.....; In.A.I.,1 I 11,2_1.11. 4 6 ti.1,41 4 I I 1 II' talj * .0.1 I 4 (420... 4 I 0 It

iii ..:11)( : II I '4," 1 I " 1 04 11111 11 ill 41. I 11,0i I 4. : .: .,-. i * I s lls : 1141 4

1.4, 11 -1111.. I 4' o i 'L t 4 :-4/:,. -.. : I o

I .0 0.40,0.0_. :AI I. I I I .-1 lb t 1 . 1 -I 11 1 : t.I - .. --4 V , I1

4 I+ , .4 ...1.11, 111..."6 6 0.1 4*

...0 '16. e ....It 4 ...: 0. 4 0 4_1 0.0; ..., .4 ,... 1 .',.

1 4 , I 4; 1401 It o a, 1 4 1 i 1 1 I 51 4 3 n_ . 1 11.1.1.5_4

' 41511 I "1.4 011 (4 **I _1....II 6, 1 It) . -.. i.

,41 I 4 I 141 il

i_t. , *III 1... 1114 ....0 0 4 I 11.4 PL V 1 4 4 4...1 4 1 4I1 0

. g 4 111 tiL,2141 4 / .0 .01 li : 0 1 41

4 o f I I I 1 i ..!_vo I. I 1 III 00 II 1 I .0: 1 ...1 4 1 4, 11 : 11

.114 5 tit I 1 .01 I 11,4 41t., ' ihzihi...1 11 101 I 1 . 4 1 r/ 1+0 I 1 : 1 .:1111

.6 I : I . I I : t $11..4 I' 'I 6 6 .411, I I:1 4, 4 - 0 I 1_

: ... L t .. I I,4 t. *boos ti 0 0 o/out

II sr o - tot . 1 o . 114 4 ...0

1 . I 4 I 041 - o : t th r 1 1. 4 LI ' I '. :IV 0,,, .1.1.: 0 I .. 1 : 4

tor.o ... 4411 1 40. III II : ..4 : II 4 4. 111..1 1 4 .114 I 1.

I 4 IDI 4. 0.:I ti L101II'''. / ".4 : 0 1.' 1 i i 1 ist 4 .- I

4 .... , :11 1 I .4 4 4 1 I 1 : o 1 101 ..: I, Iii ,40 3 4, :i I 1 41_1 I 4

1 1 .111 14 - 1 111.1, 111 6 4 .a...L.' i 6 1 t 3 1 .4 : .,4I IV

4 0I 11,11 " 1 Il 4 A : L I t 1 0 1 1 OW 1 1 4 4 4 I 0 4 0 0

I 4 10 I 4k 14(..f-V I 1. Sikril 1, 1111'1 I f 4 : O it, I' 1 0 4 I I _. 111' 1

I 1 1 1 4411 1;1/ 'I 11 1 4 1 4 4 4

.1 'i ..4_1 W': 4 ' 5 5 1 SI 0 , 0 I 11:_

I ..44.1 1 0 L__ 1 4 4 1 VII ' 4 1 I 0 : 4 4 .4, 4 4 4 1 I I

1 11/16 11 11 ' 1 41 -_. 1111. 111,1.14 . II :10, 05 14. 1

: 1 1 .6 0 1 4 0..-11 ' 1 .4/ " : 6 1 1 1 1 it 1 1 4 1 li)t 1

: ,1 11 Uk..6.411...:-4 4 44)9 ' , 1 I ' 1.1 1 1 1 13, is : 111: II,I

*1 *AI 4 5 1 0 1 1 .6 1 , 1 4 I 4 ' +1.4 4 I I I 0.i(_, 1 I : 1 1 VI ' 4 0 I; I 0 i

4,0 to

i,.bjl 111 '211 tI Ar.:1 -11 t 4. ft * S 14.0 11 1 . at

11 1 .11 11 111 11*

:1 4.11 61 4 414 3' 11 43 /1 ILLLS 4 I 0 ,:i !AI 4,, . 1 wt.. - 1, 4 g , 1 11 1 34,..1 1.

II 1 -111 .4) , a 1114 .... 4 - 1 I ... I I I. 11 1 / : 61 . 11 1 '

%II ti.11'. jik 1 4 .1t,L1 It,_._ 11 (t .... : I .14 : 1

0- 41 5. _.oloo 1_7.1 4) u4i ... 1.' 4 .1.- 14111141L 4, .1 141 . I 1 ..*4 ... ..!_t 41

. is 4II I _A , 1 ) I 1.4 1 .41 ... . .5 5 ; 0 61oi4: 4 111 1 - 6 1 0 4 ... .11 . 0 il1 4 111 . 41 1 0 0 ' 1

te shielding) had a measurable effect on behavior. He found that underlded lights the students awed in significantly fewer off-task behavion.

Summary

Rosana to a number of non-viesial effects of light on people. Of these effects tinsuntan ì -44 and the control of rickets are two that are quite well understood. Otherimportant (but perhaps less well known) effects include: seasonal affective disorder (SAD),synchronization of a number of physiological rh and the control of infantilejaundice. 1 artificial JI systems provide 1 tunination equivalent to on'y one ortwo percent of intensity " day restarch also -- that variations in the qualityof artificial light in classnxans can have an effect on " ""tSo

The action spectra d importance to hunizns ranges from 290 to 770nm. Skin g and vitamin synthesis occurs in the range of 290 to 315 nm.Tanning or gmentation of the skin and reduction of dental caries occurs in response tolight in the 2.10 to 403 nm. Vince is most sensitive to light in the 500 to 650nm range (yellow-gmen light). Bilinibin degradation OCCUrs ill response to light in the 400

necessary fir all of these biological functions. Full-to 500 mn range (blue light). Natural light ;Provides the rocItexas Iceneitte guymindisInbuaitiz

provides substantially all of the spectial energy distri ,on although light levels are muchlower than daylight levels. The spectra of incandescent, cool-white fluorescent, and highpressure sodium vapor light sources appear to fall short of covering the entirephotobiologic action spectra of importance to hinnan beings.

For people who are outdocas for a significant wt of each day, the quality of indoorlighting to which they are exposed may be of little consequence. Their needs for naturallight stimulation may be adequately met.

But for people who spend almost all of their time indoors, and with outdoor exposurelimited to morning and evening light, there may be a need for artificial lighting that issupplemented with light stimulation in the spectrum aim of enemy deficiencyespeciallyblue (440 to 440 nm) and ultraviolet in the no to 400 nm zange (UVA and UVB).

Building on this background of zesearch, a classzoom-based study conducted in Albertafrom 19111 to 1985 identified what to be the very significant finding that elementazystudents receiving trace amounts ultraviolet light developed fewer dental caries thanthose students of a similar age and background that did not receive ultraviolet light

The research reported in the literature also suggests that exposure to differentlighting systems may lead to differences in the way children developlight may also havean effect on the overall growth and development of children.

ChsENsty 3

Study Design

The current study was designed around a pretest-posttest, multiple-group model andintended to plow that:

Different light types (full spectrum fluorescent, full fluorescentwith UV enhancement, cool-white fluorescou, and , 21, Immure sodiumvapor) have no differanial effects on Division II stadiums' dental histories,growth and develwment histories, vision histories, scholastic achievementhistories, or attendance histories when compared over a two-year period.

Study Sltes

Eve sngly sites were selected and include&

Site 1 (high pressure sodium vapor lighting): a school lit with indirect high pressuresodium vapor (HPSV) lamps. High pressure sodium vapor lamps mounted in indirectfixtures were used duoughout the claswooms in this school. Natural light fromwindows modified the color characteristics of these lamps somewhat.

Site 2 (full spectrum flumescent lighting without UV enhancement): a school lit withfull spectrum fluorescent lamps mounted in conventimal fixtures. The original cool-white flmescent lamps in this school were replaced by Vita-Lites distributed by Duro-Test Canada, Inc. The white baked enamel fixtures and the Plexiglass lenses wereunmodified.

Sites 3 and 5 (full spectrum fluorescent lighting with UV enhancement): two schools litwith full spectrum fluorescent lamps with UV enhancement and mounted in fixturesthat contained aluminum reflectors and aluminum grid diffusers or open egg-cratediffusers. The original cool-white fluorescent lamps in these schools were replacedwith Vita-Lites manufactured by Duro-Test Corporation. The white baked enamelfixtures were rtmdified by adding polished aluminum reflectors inasmuch as thereflectance of the white baked enamel fixtures is only 5 to 10 percent in the UV bands(Bickford, 1981: 19-17). Because plastic lenses and diffusers also filter out UVradiation, several alternatives were followed to ensure that the maximum amount of UVwas allowed to reach desk top level. Wrap-swum:1 Plexiglass lenses in some fixtureswere removed because suitable substitutes were unavailable. Where lay-in Plexiglasslenses were in use, they were removed and replaced with aluminum grid diffusers.(The grid apertures were approximately one centimeter square.) Where open egg-cratestyle diffusers were available, they were left unaltenul.

Sites 2, 3 and 5 were lampod with Vita-Lite fluorescent law for several reasons.First, the nianufactiner's specifications indicate that these laws simulate sunlight (theyprovide a balanced spectnmi including controlled amounts of ultraviolet =Nation in themid and near UV range (280 - 400 nrn). Second, these bimps are the same as thoseused in the first study reported by Wohlfanh (1986) and Hargwaves and Thompson(1989). Thild, Duro-Test Canaa, Inc. provided sufficient lamps at no cost to thestudy to relamp experimental areas at Sites 2, 3 and 5 (approximately 2,000 lamps).

25

Site 4 (cool-white fluorescent lighting): a school lit with cool-whim fluorescent lamps.Cool-white fluorescent lamps mounted in white baked enamel fixtures equipped withPlexiglass lenses were used throughout the school.

These particular study sites were selected fix a variety of reasons. One of the mostimportant was project costs as low as possible. One way of keepingcosts low was to use existing wig systems wherever possible. Site 1 was selectedbecause it was a new facility it was fully equipped with high pressure sodium vaporlamps in all classrooms. Site 2 was selected as a result of an informal request to AlbertaEducation for full spectrum lamps to replace existing cool-white fluomscent lamps. Sites3, 4 and 5 were selected because of the willingness of the district to permit three of it'sschools to be involved and further to permit the modification of the lighting systems in twoof it's schools to include new reflectixs and diffusers in the fluorescent fixtura andreplacement of existing cool-white lamps with full-spectnnn lamps manufactured by Dino-Test Corporation. It was further agreed by all of thedistricts and schools that the selectedschools would be a part of the study for two a and for that period students wouldremain in those classrooms equipped with .; systems.

"Table 3.1 serves to summarize the illumination and UV levels at each site.

A preliminary comparison.of-all classroom UV levels measured to date indicates thefollowing:

1. Aluminized reflectors enhance ambient UV emission levels from four to six times overconventional white-painted fixtures. (A comparison of the UV reflectivecharmeristics of a number of surfaces is presented in Table 2.2.)

Fixtures without lenses or diffusers create significant UV "hot-spots" which are about1.3 times pester than average ambient levels.

3. The Vita-Lite full-spectrum Imps emit UV levels approximately 10 times higher thanthat of high pressure sodium vapor lamps and approximately four times that of cool-white fluorescent lamps.

4. The aluminum grid diffusers create the most uniform UV levels throughout theclassroom.

5. High pressure sodium vapor lighting systems emit insignificant quantities of UV.

Study PopulationThe study included all of the Grade 4 students enrolled in each of the schools at the end ofthe 1986-87 school year. These students remained in the study to the end of the 1988-89school year. Grade 4 students were selected in the 1986-87 school year for three reasons.First, this is the age group that was involved in dm first study completed in Alberta from1981 to 1985 which looked at the effects of light on development of dental caries. Second,elementary students usually remain in homemoms for most of the school day and for thatreason it is easier to retain them under the selected lighting system for the duration of thestudy. Third, students at this age are on the brink of, or already imdergoing, significantphysiological development (including the eniption of permanent teeth). Only thosestudents remaining at their respective study Sites for the duration of the study wereexamined in the final data collection phase. The intent was that by starting with all of thestudents in Grade 4 in June 1987 at each school (40 - 50 students), at least 20 studentswould remain in each group to the conclusion of the study.

14 26

Table 3.1. Measured light in classrooms. (Averages of several participatingclassrooms were taken in each school. Measures were taken at the end ofthe study and reflect the minimum levels of Uv exposure riming the study.)

site(Light 1 We)

U VA315-400 nm

UWicm2

UVB2/0-315 um

ItWA:m2

U V C100-280 nm

nWfcro2

Illumination(Thotomenic

ningPLux"

Site 1 0.2 1 0.00 0.00 250-540Wish Pressure Sodium Ave (0.19, 0.23) Ave (0.0, 0.0)Vapor Lamps

Site 2 1 .01 0.00 0.00 300-9 00Full Spectrum Fluorescent Ave (134, 0.48) Ave (0.0, 0.0)WV inhilited)

Site 3 7.20 0.30 000 280-450Full Spectrum Flucrescem Single mean= Single measure(UV enhanced)

Site 4 0.8 7 0.07 0.00 250-540Cool-wkite Fluorescent Ave (0.60, 1.15) Ave (0.0, 0.13)

Site S 5.18 0.1 8 0.00 220-450Full Spectrum Ruoresvent Ave (332, 5.63, Ave (0.11, 0.19,(UV enhanced) 6.40) 0.24)

* Variations in readings at the upper end ot the range are the result of substantial amounts of flannel lightentering the classrooms. The lower end of the range reflects the minimum amount of light available inthe classrooms and may be mat representative of the emissions from the classroom lighting systems.

Ultrayiglet (UV) Safety LiminUv radiation safety limits fa an eight (8) how period are as follows: (Phillips, 1983:4):

UVA (315-400 nm): 1000 pW/cm2UVB + UVC (100-315 nm): C.1 11W/cm2

maigagingkaigum1. UVX-36 broadband ultraviolet meter menufactured by Ulaiviolet Products Ltd. This meter was used to

pad= broadbmd UV measurements over various areas of each mom to assess the uniformity of theultraviolet levels throughout the classroom. A series of measurements was performed az selectedlocations. Direction of measurements included: (i) horizontally directed towards windows (sunlightcontribution); (ii) horizontally directed away from windows WV classroom liglaing-ambient); and (iii)venically directed towards the classroom fighting (maximum IN lighting contribution).

2. IL-791 Spectroradiometer System manufactured by International Light. This NBS tremble calibratedsystem was used to conduct natrow-band (5 am bandwidth) measurements of ultraviolet wave band from200 to 400 ammeters. A locetion in the classroom corresponding to an "average" UV exposure levelwas chosen. At this location, meatalleMIMIS were taken horizontally in * direction perpendicular to thesunlight flux entering the clamoom. Another sat of measummens was performed vertically towards thelighting system. The first set of data correspond to the average ambient classroom UV levels, while thesecond set we useful for ccmparisons of the different light(leas/reflector combinations.

3. Model 40X Optometer manufamereed by United Detector Technology Inc. Visible light fihuninance wasassessed at the chosen "average UV position" in du chusroom. A range of illuminance was found via a360° horizontal wen about the chosen position.

157

Data Collection

A number of different kinds of data were collected from the students includinv.

age,sex,nutrition histories,fluorine levels,dental histaies,attendance histories,maul health and powth anddevelopment histories,scholastic achievement histories,sldn types, andvisiar histories.

Controlling for Bias and Contaminating Variables

Inasmuch as the study sites were selected on the basis of either existing lighting systems orthe ease of converting to new lighting systems, the ability to randomly select schools andsubjects was lost. Nevertheless, several things were done to reduce the risk ofcontaminating variables in the study.

1. The lighting systems were changed during July and August so that they were in placewhen children returned to school in Seyember, 1987.

2. The nature of the study and the potential outcomes were not discussed with anyone inthe schools. In view of the rather limited research in circulation on the subject of non-visual effects of light on people, the probability is low that teachers and suidena couldhave guessed the nature of the anticipated study outcomes.

3. Administering achievement tests and maintaining attendance records were taskscompleted by teachers and no external researchers intervened. To students, it wouldhave been difficult to discern that some of the information collected by teachers duringthe two years of the study was earmaiked for study purposes.

4. Data were collected in =let to assess the comparabilityof the student groups and all ofthe pre-test data collection was useful in this regard. The nutrition data was collectedand examined in conjunction with the dental data but it was also used as a proxy forsocio-economic and cultural data. It was believed thatdifferences in total daily calorieintake or differences in potein and sugar intake couldpoint to socio-economic andcultural differences in the groups.

5. The health, nutrition and dental research= were not made aware of the school lightingtypes. Experience suggests that unless one is very knowledgeable about tingsystems, they would not have discemd the type of lighting present in the 6; i trentschools just by a visual inspection. Indeed, in some of the schools the test lamps wereonly in classroom areas and would not have been visible to researchers examining ortesting students in specially assigned rooms.

6. Statistical routines designed to overcome the inability to randomly assign subjects tostudy groups were used in some of the data aneyses.

16S

Dental Histories

Procedures developed during the 1981 to 1985 study (Hargreaves & . 1989)wen used in this study and were canied out by the same research= involved the earlierstudy. The procedures used in that first surdy (and repeated in this study) are as follows.

The nostic standards were based on criteria similar to previous surveysby the authors [Hargreaves and Chester, 19731. Portable *Nal

chain were used, and illumination was achieved by Anglepoise lamps with100 watt bulbs and nantral daylight, by the dental chair in well-litconditions. Disposable mirrors and rep sickle-shaped probes wereused. Examinations were carried out at the respective schools withoutprecleaning the teak however, the teeth were air-dried with a chip syringebefore nomination. Caries were assessed as previously described by theauthors [Hargreaves and Chester, 1973) which fulfilled WHO standards[19771. Teeth and tooth surfaces were recorded to complete DEFT andDEMS scores. The caries experience in the primary dentition was notmeasured because of the subjects' age, most primary teeth havingexfoliate& Degrees of caries were assessed based on a modification of thecriteria Zy Moller 119661 ; 'sound': no clinically deteaable defect detected inthe tooth swface; 'caries 1`.7 minimal enamel deect, detected by a 'catch'with a sickle probe or a decalcrication without enamel penetration; 'caries2'; marked involvement of enamel andlor dentine with detection by a'sticking' siclde probe or obvious tooth loss from caries; 'caries 3': severetooth loss from caries with probable pulpal death. A tooth was considerederupted of any pan of the crown was observed in the mouth (Hargreaves,1958). No radiographs were used in order to avoid excessive use orduplication of rodiogruphs with this age group. Oral hygiene was assessedby the modified debris inda of Me and Silness [19641...Ater- and intra-examiner reliability was maintained a: a level of Kappa 0.8 [Cleaton-Jones,et al., in press].

Each child's teeth were examined twice during the study, first at the end of the 1986-87school year (as students were preparing to leave Grade 4), and again at the end of the 1988-89 school year as the students were leaving Grade 6. For each child all erupted teeth (i.e.,all teeth where the crown was visible in the mouth) were examined and both decayed,extracted and filled teeth (DEFT) and decayed, extracted and filled surfaces (DEFS) wererecorded. The examination assessed degrees of caries as follows:

Sound no clinically detectable defect in the tooth surface.

Caries I (a) minimat enamel defect, detected by a "catch" with a sickle probe, or (b) adecalcification without enamel penettation.

Caries 2 (a) marked involvement of enamel andior dentine with detection by a "sticking"sickle probe, or (b) obvious tooth loss from caries.

Cada. 3 severe tooth loss from caries with probable pulpal death.

General Health and Growth and Development Histories

Information about general health and the growth and development of students was collectedby a Registered Nurse and included the following:

age at onset of menarche,height,weight (measured with a Seca personal scale accurate to0.5 kilograms),body fat (measured with a Lange Skinfold Caliper calibrated in millimeters),vision,skin types, andother health data.

Vision Histories

Vision histories included

uncorrected distance vision (left eye, right eye, both eyes together) which wasmeasured by means of a standard 20-Foot Snellen Chart based on a visual angle of oneminute (supplied by Avenue Medical Supply Ltd.) and illuminated to normal classroomlighting levels, and

use of glasses.

The Snellen Chart was calibrated from 201200 (poorest vision) to 20/10 (best vision). Thedenominator in the ratio was recorded and included in the data analysis. Based onHelveston, et al (1985: 346-355), normal distance vision was taken to be 20130 or betterand abnormal vision taken to be 20/40 or worse.

Attendance Histories

To ckvelop attendance histories for each student, monthlyattendance was zecorded in half-day inczements and compared to the maximum number of days the schools were open.

Scholastic Achievement Histories

Scholastic achievement histories were developed by administering the complete CanadianTest of Basic Skills: Form 5 (Level 10 for tha& 4 and Level 12 for Grade 6). Thecompleted data set included 15 measures: vocabulary, reading, language (four tests and asubscore), work study (two tests and a subscore), mathematics (three tests and asubscore), and a total score.

Nutrition Histories

Nutrition histories were collected on two occasions during the study by means of theThree-Day Personal Daily Menu Diary. On each occasion a nutritionist instructed thestudents with respect to maintaining an accurate diary. When the diaries were completed, anutritionist discussed them with students individually and added any details that had beenoverlooked.

Skin Types

With the notion in mind that skin pigmentation may lead to differential effects of ultravioletlight on students, student skin types were noted and recorded. The classification of skintypes was based on a system reported by Pathak (1985: 328) and includes the followit., six

tYPes:

Skin :we I. These individuals have very fair skin, zed or blond hair, blue eyes, and arcoften freckled. They burn easily, develop painful sunburn, awn peel, and do not tan.They are most often ofCeltic background (e.g., Irish, Scottish, W )

Skin° e IL These individuals have fair skin, blue or hazel eyes, zed or blond hair, and areoften freckled. They 63 not tan well, but upon (4' 4 exposures, "harden" their skin and=are a slight tan. The freckled skin becomes and the non-freckled skin remains

Skin type IlL These individuals have a light tan or medium color skin with blond,bnmette, or light pigmented hair. They burn moderately, then tan moderately anduniformly. They acquire an average or good delayed tanning reaction with two or threeexposures and do not burn intensely on subsequent sun exposures.

Skin ope IV. These individuals are lightly tanned awl can tolerate the sun a !ittle betterthan skin types I and U. Um:posed skin it, light brown or olive. Eyes and hair are mostlikely to be dark. The group 4:sichKles pigmented Caucasoids, Red Indians, Chinese,Japanese, and people from the Mediten-arsean (Italians, Greeks, Spniards). In the summerthe facial sldn color changes from light brown to olive or medium brown.

Skin t e V. These are brown-sldnned people who can tolerate the sun quite well. Theeyes and hair are deep brown or black. This group included Mexicans, East Indians,Egyptians, Malaysians, Pueno Ricans, and other Spanish-speaking people.

Skin type VI. These are markedly pigmented individuals. The skin is brown to black incolor. These individuals generally neves burn, but become profusely dark skinned afterexposure to the sun. This group includes African and American Blacks and AustralianAborigines.

Data Processing

The collected data were translated into electronic form, checked for accuracy, andtransferred to the computer at the University of Alberta. Data summary and statisticalanalyses were performed under the direction of Dr. Gordon Thompson, Faculty ofDentistry, University of Alberta. Analysis of attendance, achievement and health-relateddata were completed by researchers within Alberta Education.

3r

ChspilsT 414

Study FindingsThis current study was designed as a replication of a study carried out from 1981 to 1985wherein it was found that trace amounts of ultraviolet light in classrooms were linked to: (a)reductions in the development of dental caries, and (b) improved attendance. Inasmuch asfindings from other research projects seemed to point to the conclusion that differentlighting systems may have differential effects on people, a number of diffinent kinds ofdata were collected from the students in this study, including:

age,SCX,nutrition histories,fluorine levels,dual histories,attendance histories,gencal health and growth and development histories,scholasdc achievement histories,sldn types, andvision histories.

FOT the most part the data sets that were collected were intended for urc in interpreting thefindings of the dental and attendance components of the study. Accordingly, the dental andattendance components of the st-Cy will be reported and discussed first and then followedby the remaining data sets.

A Description of the Study Population

The students involved in this study were located in five different schools. In order todetermine the comparability of students and sites, information about students ages, sex,nutrition, and fluorine levels in water were collected and analyzed.

Age

The average age of all of the students involved in the study was 1102 years as of June 30,1989. Table 4.1 describes the range of ages of the students at the five study sites. Nosiwrificant age differences were found among the study populations.

S e x

The composition, by sex, of the study group is described in Table 4.2.

Nutrition

Nutritional data were collected for three purposes: (a) to serve as a proxy measure inestablishing the comparability of student populations at the five sites, (b) to determine ifnutrition deficiencies might have an impact on development of dental caries, and (c) todetermine if nutrition deficiencies might have an impact on overall health.

Nutrition histories were collected by means o f aThree-Day Personal Daily Menu Diary..Site specific dam with respect to daily calorie, protein and sugar intake are presented in

32

Table 4.1. A two-tailed test comparison of the age of study w&-ipmts asof June 30,1989. Levels of significance beyond the 0.05 level are in bold faced italics.

Site

MeanSt Dev

N

Site 1 Site 211,97 12.040.23 0.5238 35

Site 312.1120.3644

Site 411.9802743

Site S12.060.3656

Site 1 (HPSV)

Site 2 (FS)

Site 3 (FS + UV)

Site 4 (CW)

Site 5 (FS + UV)

11.97

12.04

12.02

11.98

12.06

0.23

052

0.36

0.27

0.36

38

35

44

43

56

0.723 0.750

0.191

0.178

0.610

0.580

1.463

0.197

0546

1.251

Imm.a

Table 41. Sex of study participants.

Site TotalNumberof Males

Percentof Total

Numberof

FemalesPercentof Total

Site 1 (HPSV) 41 24 0.585 17 0,415

Site 2 (FS) 37 22 0.595 15 0A05

Sites 3 and 5 (FS + UV) 147 74 0.503 73 0.497

Site 4 (CW) 58 28 0.483 30 0.517

Total 283 148 0.523 135 0.477

Table 4.3. Table 4.4 reports data for the combined sites receiving UV supplements andthrse not receiving UV t lement& No significant differences were found in the dailynutrition of studarts at the e sites or between the UV group and the non-UV group.

Inasmutir as no significant differemes were frxmd in the ages or daily nutrition of studentsat the different sites, the conclusim is supported that the students at the five sites are quitecomparable. The argument is that substantial differences in culture or socio-economicfactors might have altered either the global daily calorie intake or the composition of the diet(e.g., differences in protein ix' sugar intake).

Fluorine Levels

There is abundant evidence that the addition of small quantities of fluorine to drinking valutaor the local application of fluorine to the teeth of growing children greatly reduces theincidence of caries. The most common practice is to add fluorine to the drinking water tobring the fluorine content up to 0.5 - 1.0 pans per million.

Public water supplies at all of the study sites were adjusted to the range of 1.0 to 1.1 pansper million of fluorine.

Study Findings

The purpose of this study was to test the null hypothesis that light dces not have non-visualeffects on students in classrooms with different li,ghting environment& This section servesto test the null hypothesis in a number of ways.

Dental Effects

Those students receiving trace amounts of UV emitted from Vita-Lites (Sites 3 and 5)developed signifimmtly fewer dental caries than students under other lighting environments(Sites 1 and 2). These results are presented in Table 4.5. Table 4.6 summarizes thefindings by comparing the sites receiving ultraviolet supplements with the sites that did notreceive the UV supplements.

Snidents at Site 4 (lamped with cool white flucrescent lamps) were dropped from the studywhen it was found that a high percentage of these students had fissure sealants applied totheir teeth during the course of the gudy. Because fewer students at the other sites receivedfissure sealants (fissure sealants were readily detectable by the dental examiners), theresults aze reported with students receiving fissure sealants both imluded and excluded Asmay be emected, the beneficial effects oIIJV radiation are understated when the studentswith fissure sealants are included.

When students receiving fissure sealants are included in the analysis, students in theControl Group had an average of 039 decayed teeth (or 0.95 decayed surfaces) more thanthe Treatnznt Group. When students receiving fissure sealants are excluded from theanalysis, the Control Group had an average of 0.79 decayed teeth (or 1.36 decayedsurfaces) more than the Treatment Group. Complete reversals (clinical signs of earlycaries at the initial examination not being detectable at the final examination) were observedfor some tooth surfaces, specifically in the Treatment Group receiving the UVsupplements.

The findings from site to site with respect to reductions in dental caries development as aresult of ultraviolet radiation are apparently not influenced by diet based on macro-nutrient

Table 4.3. A two-tailed r testintake. Levels )

of daily nutrition levels-calorie,significance beyond the 045 level are in

Daily Calorie Intake

MeanSt Dev

Site

Site 1 1772.0 658.3 19

Site 2 1697.4 527.8 28

Site 3 1642.6 450.1 37

Site 4 1577.3 406.8 36

Site 5 1736.2 497.1 37

Site 1 Site 2 Site 3 Site 4 Site 51772.0 16974 1642.6 1577.3 1736.26583 527.8 450.1 406.8 497.1

19 28 37 36 37

0.402 0.751 1.147 0204

0.434 0.979 0.296

0.642 0.837

1.475

Daily Protein Intake in Grams

Site 1 She 2 Site 3 Site 4 Site 5Mean 59.4 67.4 61.7 60.7 63.4

St Dev 24.6 24.4 18.0 17.0 21.5Site N 19 28 37 36 37

Site 1 59.4 24.6 19 - 1.071 0.343 0.189 0386

Site 2 67.4 24.4 28 - 1.031 1.228 0.677

Site 3 61.7 18.0 37 - 0.245 0.375

Site 4 60.7 17.0 36 - 0.603

Site 5 63.4 21,5 37 -

Daily Sugar Intake in Grams

MeanSt Dev

Site

Site 1 107.2 51.0 19

Site 2 102.9 39.9 28

Site 3 101,3 50.9 37

Site 4 102.9 41.6 36

Site 5 106.6 36.5 37

$

Site 1 Site 2 Site 3 Site 4 Site 3107.2 102.9 101.3 102.9 106.651.0 39.9 50.9 41.6 36.5

19 28 37 36 37

0.302 0.404 0.312 0.045

0.142 0.004 0.378

0.144 0311

0.403

MEW

Table 4.4. A two-tailed I test compadson of daily nutrition lorelscakrie, protein, andsugar intakefor UV enhanced sites and um-UV enhanced sites. Levelsa significance beyond the 0.05 level ate in bold faced italics.

Daily Calorie Intake

Non-UVSites

Mean 1662.4St Dev 513.7

Site N 83

UV Supplement Sites 1689.4 473.3 74 0.341

Daily Protein Intake in Grams

Non-UVSites

Mean 62.7St Dev 21.6

Site N 83

UV Supplement Sites 62.6 19.7 74 0.030

Site

UV Supplement Sites

Daily Sugar Intake in Grams

MeanSt Dev

104.0 44.1 74

Non-UVSites103.942.883

0.010

25 6

Table 4.5. Effects tf supplemental ultraviolet light in classrooms co thedevelopment of dental caries in elementary students in Grades 4, 5 and6.

Site

Incremental Inarease inCaries 1911749

Students with FissureSealants Included in

AnalysisDEFT* DEFS*

Incremental Increase inCaries 198749

Students with FissureSeakuab Exelsulei from

AnalysisDEFT DEFS

Site 1 1.13 1.45 1.30 1.72

High Pressure Sodium VaporLamPs

Site 2 0.68 1.18 0.70 1.33

Full Spectrum Fluorescent(UV inhabited)

Site 3 0.42 0.40 0.22 0.14

Full Spectrum Fluorescent(LN released)

Site 5 0.21 0.34 0.19 0.19Full Spectrum Fluorescent(UV named)

DEFT means &rayed, extracted, or filled teeth. DEFS means decayed, extracted, or filled solaces.

Students at Site 4 (Cool white lamps without UV) were eliminated from the data analysis when it wasdiscovered that a high percentage of these smdents had received applicationsof fissure sealants during

the study paiod.

26 37

# lemental ultraviolet light in classroomcaries in elementary students in Grades

cm4, 5

Table 4.6. Surmnarized effects of -the development of den #and 6.

Incremental Increase inCaries 193749

Students with FissureSealants included in

Analysis

Incremental Increase InCaries 193749

Students with FissureSealants Excluded from

AnalysisSite DEFT DEFS* DEFT DEFS

Non.UV Control Group 0.91 1.32 1.00 1.53Average

UV Treatment Group 0.32 0.37 0.21 0.17Average

Differences (fewercavities)

0.59 0.95 0.79 1.36

* DEFT means decayed, muscled, or filled teak DEFS means decayed, extracted, or filled =faces.

Site 4 (Cool white lamps without UV) was eliminated &mu the data analysis when it was discovaedthat a high percentage of the students had received applications of fissure sealans during the slimlyperiod.

assessment or by fluoride levels of the drinking water. As well, all children involved in thestudy had equal access to over-the-counter Feventive dental health piothicts and more than95 parent of dentifrices available in the marketplace have a fluoride component of 1,000 to

19250 IWO-

These fmdings support the view that the ultraviolet light component of the classroomlighting systems may best explain the differences in caries increments found amongsmdents at different sites.

Conclusion. On the basis of an analysis of dental records for the childrenlocated in the different lighting environments examined in this study, thenull hypothesis was rejected-4ight does have an effect on the developmento f dental caries.

Attendance Effects

To develop attendance histories, monthly attendance was iecorded forawl student in half-day increments and then divided by the maxim= number of days the schools were open inorder to derive the parentage of attendance. The findings am presented in Table 4.7.

A number of significant differences in attendance are to be noted. Site 2 (full spectrum)and Sites 3 and 5 (full spectrum with UV supplement) had significantly better attendancethan Site 1. The difference amounts to approximately 3.2 days per yea. One mightconclude that an absence of 3.2 days is about the length of time that it would take to recoverfrom a severe cold.

Conclusion. On the basis of an analysis of attendance records for thechildren located in the different lighting environments examined in thisstudy, the null hypothesis was rejectedlight does have an effect onattendance rates.

General Health and Growth and Development Effects

A number of pieces of information pertaining to the general health and to the growth anddevelopment of the students in the study ?pups were collected by a Registered Nine withthe expectation that these data may be of use in interpreting findings with respect to effectsof light on students. An analysis of the information provides the following findings.

Age at onset of menarche

A relatively small percentage of the overall female study population reached menarcheduring the course of the study (17.8 percent), however the distribution was not entirelyuniform. To establish an expectation of the number of girls that should have reached theonset of menarche, the results of a completed study of 1829 girls was provided y theEdmonton Board of Health. Table 4.8 presents the data collected in this study.compared to the study of 1829 girls, the higher than predicted incidence of onset ofmenarche at Sites 2 (full spectrum) and Sites 3 and 5 (full spectrwu with UV su lement)and the lower than predicted incidence of the onset of menarche at Site 1 is t tonplain. One factor common to all of these sites is lightthe higher incidence occurred inscols lit by Vita-Lite full k k kk lamps and the lower incidence was in the school lit byhigh pressure sodium vapor

Table 4.7. A two-tailed :test comparison of rates of attendance (attendance ratesrepresem the actual attendance divided by possible days for the periodSeptember 1987 to June 1989). Levels of significawe beyond the 0.05level are in hid faced italics.

Site

MeanSt Dev

N

Site 1 Site 20.943 0.9620.051 0.01667 36

Sites 3150.9590.038

136

Site 40.9590.040

56

Site 1 (HPSV) 0.943 0.051 67 - 2.780 2.260 1.933

Site 2 (FS) 0.962 0.016 36 - 0.707 0.497

Sites 3 and 5 0.959 0.038 136 - 0.000(FS + UV)

Site 4 (CW) 0.959 0.040 56

Table 4.8. Incidence of erls reaching the onset of menarche. Statisticallysignificant differences beyond the 0.05 level, as determined by meansof Ohi Square tests, are shown in bold faced italics.

Site

TotalNumber of

Girls

NumberReachingMenarche

AverageAge at

Onset ofMenarche

PredictedProbability

ActualPercent

ReachingMenarche

Site 1 (HPSV) 17 2 11.763 0.218 0.118

She 2 (FS) 15 7 11.729 0210 0.467

Sites 3 and 5 (FS + 73 12 11.069 0.095 0.164UV)

Site 4 (CW) 30 3 11.238 0.118 0.100

Totals 135 24 11.340 0.135 0.178

* Probabilty based on a study of 1,829 girls reported by the Edmontrm Board of Health.

2940

Height Gains

One measure of general health which was collected at the study sites was gains in height(measured in cendmeters). This infannation is &splayed in Table 4.9.

The smallest gains in height were made at Site 1 (high pressure sodium vapor) while thegreatest gains were made at Sites 3 and 5 (full spectrum with UV t). Indeed, thegains at Site 1 were significantly less than gains at Sites 3 and 5 ( spectrum with UVsupplement) and Site 4 (cool-white).

Weight Gains

Weight gains were recorded for smdents at all sites with the weight gains measured inkilograms. These data are summarized in Table 4.10.

The greatest average weight gain occurred at Sites 3 and 5 (full spect:rum with UVsupplement) and these gains were significantly greater than those at Site 4 (cool-white).Arguably, light could be a clear factor in this instance inasmuch as both sites are located inthe same community and other factors are more or less equal.

Body Fat Gains

Body fat was measured with a Lange Skinfold Caliper calibrated in millimeters. Changesin body fat were calculated by subtracting the pre-test midis from the post-test results.

Increases in body fat during the period of the study were lowest at Site 4 and highest at Site2. Differences that are statistically different beyond the 0.05 level are displayed in Table4.11.

Conclusion. On the basis of an analysis of health and general developmentrecords for the children located in the different lighting environmentsexamined in this study, the null lopothesis was rejectedlight does have aneffect on health and general development. Specifically, significantdifferences were found to include; age at the onset of menarche, heightgains, weight gains, and gains in body fat.

Scholastic Achievement Effects

Scholastic achievenunt histories were developed by administering the complete CanadianTest of Basic Skills: Form 5 (Level 10 for Gra& 4 and Level 12 for Grade 6). Thecompleted data set included 15 measures: vocabulary, reading, language (four tests and asubscore), work study (two tests and a subscore), mathematics (three tests and asubscore), and a total sem.

As is the case with many research projects, it is often necessary to study groups as theyarethe subjects cannot be matched or assigned at random. Such was the case with thisstudy. In cases like this Kerlinger (1964: 347-348) is of the view that Analysis ofCovariance is a preferred smtistical procedure"it tests the significance of differencesbetween means of the final experimental data by taking into account and adjusting initialdifferences in data." Ferguson (1971, 288) describes the Analysis of Covariance as a"statistical, rather than an experimental, method [that] may be used to 'control' or 'adjustfor' the effects of one or more uncontrolled variables [added by author]"in this casedifferences in pretest scores.

30 41

Table 4.9. A two-tailed r test $ $1$ $ of height gains measured in centimetersbetween June 1987 e June 1989. Levels &significance beyond the0.05 level are in bold faced italics.

S Ile

MeanSt Dev

Site 110.23.6

N 29

Site 212.03.934

Sites 3151232.9110

Site 411.92.746

Site 1 (HPSV)

Site 2 (FS)

Sites 3 and 5(FS + UV)

Site 4 (CW)

10,2

12.0

12.3

11.9

3.6

3.9

2.9

2.7

29 -34

110

46

1.873

-2.827

0.378

-

2.151

0.127

0.772

AINNIP

Table 4.10. A two-tailed r test .1111111 son of weight gains measured in kilogramsbetween June 1987 $ $ June 1989. Levels of significance beyond the0.05 level are in bold faced italics.

Site

MeanSt Dev

N

Site 1 Site 210.2 10.94.5 5329 34

Sites 3/511.24.942

Site 49.63.547

Site 1 (HPSV) 10.2 4.5 29 - 0.558 1163 0.603

Site 2 (FS) 10.9 5.3 34 - -0323 1.230

Sites 3 and 5 11.2 4.9 110 - 2.335(ES + uv)

Site 4 (CW) 9.6 3.5 47 MM.

3142

Table 4.11. A two-tailed test +=prison of gains in body fat measured inmillimesers between June 1987 and June 1989. Levelsof significancebeyced the 0.05 level are in bold faced italics.

Site

MeanSt My

N

Site 1 Site 234 7.65.9 6229 33

Sites 3/53.96.1110

Site 40.44.846

Site 1 (HPSV)

Site 2 (FS)

Sites 3 and 5(FS + UV)

Site 4 (CW)

3.4

7.6

3.9

0.4

5.9

6.2

6.1

4.8

29

33

110

46

2.686 0.398

2.984

2.264

5.501

3.801

411=

32 4 3

When Analyvis of Covariance was applied to the dam collected by Ws=.,i Wg theCanadian Test of Basic Skills in this study, t differences beyond the 0.05 Welwere found in total achievement gains (p=o. I , F=11,423) awl in gains in lan(p=0.003, F=4.877), work study (p=0.040, F=2.826), and mathematics 001,F=5.912) between the four groups formed on the basis of classroom lighting e., highpressure sodium vapor, full specmun fluorescent, full spectrum fluoracent with UVsupplements, and cool-white fluorescent).

Site specific data with respect to total gains on the Canadian Test of Basic Skills aresunnnarized in Table 4.12. The lowest achievement gains are to be found at Site 1 whilethe greatest gains me found at Site 2 awl Sites 3 and 5.

Conclusion. On the basis of an analysis of achievement records for thechildren located in the different lighting environments examined in thisstudy, the null lopothesis was rejectedlight does have an effect on nuesof achievement.

Skin Type Effects

Corth (1984) suggested that skin pigmentatim may influence the extent to which ultravioletlight is absorbed by the skin. Skin types were recorded so that any relationships betweenskin pigmentation and development of dental caries could be explored at those sitesreceiving UV supplement& The results which were non-significant are reported in Table4.13.

Conclusion. On the basis of an analysis of skin types for the childrenlocated in the different lighting environments ermined in this study, tit wasconcluded that skin type does not modify the light effect on the developmentof dental caries.

Vision Effects

Vision histories included data with respect to:

uncorrected distance vision (left eye, right eye, both eyes together) was measured bymeans of a standard 20-foot Snellen chart based on a visual angle of one minute andilluminated to normal classroom lighting levels, and

use of glasses.

Based on Helveston, et al (1985: 346-355), normal distance vision was taken to be 20(30or better and abnormal vision taken to be 20/40 or worse.

Baseline data collected with respect to uncorrected distance vision at the beginning and endof the study are displayed in Table 4.14. Though no stadnically significant changes inuncorrected distance vision were detected, some interesting patterns may be seen in Figure4.1.

The changes in uncorrected distance vision at Site 2 (from 20(37 to 20138.7) and Site 4(from 20/47.3 to 20/48.9) parallel each other. Changes at both Site 1 (from 20/39.1 to20/44.7) and Sites 3 and 5 (from 20/39.6 to 20151.1) appear to be taking place at faster ratethan at Sites 2 and 4. There is no easy explanation for the more pronounced changes at Site1. At Sites 3 and 5, however, the more pronounced changes may be linked to the natureof the fixtures which contained the full spectrum lamps.

33

4 4

Table 4.12. A two-tailed test comparison of total achievemem gains (combinedsubsca res on Canadian Test of Basic Skills) measured in grades betweenJune 1987 and Jum 1989. Levels of significance beyond the 0.05 level arein bold faced italics

Site

MeanSt Dev

EMNIMla

Site 11.610.70

N 43

Site 22.25037

34

Sites 3/51.960.45109

Site 41.880.434

Site 1 (BPSV)

Site 2 (FS)

Sites 3 and 5(FS + UV)

Site 4 (CW)

1.61

2.25

1.96

1.88

0.70

0.37

0.45

0.40

43

34

109

34

5.089 3.008

3.737

2.101

3.901

0.976

344 5

Table 4.13. A two-tailed t test comparison of skin types in relation to the incidenceof &sisal caries. Levels of significance beyond the 0.05 level are in boldfaced italics.

Incremental Decayed, Filled Teeth (DEFT)

Skin TyPe

MeanSt Dev

SkinType 1

0.601,17

N 10

SkinType 2

0.290.61

14

SkinType 3

0.181.0438

SkinType 4

0.571.3621

Type 1

Type 2

Type 3

Type 4

0.60

0.29

0.18

0,57

1.17

0,61

1.04

136

10

14

38

21

0.729 0.986

0.457

0.061

0.805

1.118

OMIM

Incremental Decayed, Filled Surfaces (DEFS)

Skin Tyne

MeanSt Dev

N

Skin SkinType 1 Type 2

1`0 0.432.56 1.34

10 14

SkinType 3

0.081.5538

SkinType 4

0.521.7821

Type 1 1.10 2.56 10 0.720 1.145 0.616

Type 2 0.43 1.34 14 0.777 0.165

Type 3 0.08 1.55 38 0.931

Type 4 0.52 1.78 21

Table 4.14. A two-tailed t test comparison of pretest vs posit= changes inunconected distal= visionboth eyes together. Levels of significancebeyond the 0.05 level are in bold faced italics.

SitePre-Test Post-Test

t ValueMean St Dew N Mean St Dew N

Site 1 39,1 40.4 41 44.7 47.9 29 0505

Site 2 37.0 42.6 37 38.7 44.7 34 0.161

Sites 3 and 5 39.6 46.2 147 51.1 63.1 111 1.613

Site 4 47.3 58.7 58 48.9 60.5 47 0.135

364 7

55.0

52.5

50.0

5 47.5i 45.0

/ 42.5

i 40.0a 37.5

35.0

32.5

30.0

0 Site 1 a We 2 Sites 3/5 Site 4

Figure 4.1. Changes in distance vision (both eyes) as measured by a standani 20-FootSnellen Eye Chan. Normal vision is taken to be or better andabnormal vision taken to be 20/40 or worse (Helveson, et al, 1985: 346-355).

In ortkr to maximize the UV output flan lamps at these two sites two things were &one.First, polished reflectors wae placed behind the lamps. Second, where they wereobtainable, special altunimun grid diffusers were used toreplace dm Plexiglass hems. Oneschool has older fixtmes with egg-crate diffusers and these were not changed. At theseccurd school, many of the fixtures were of the surface mounted type with wrap-roundPlexiglass lenses. Because replacements could not be found, these laws were left withoutlenses or diffusers. Moreover, the ceilings were lower than in the other schools. Theresult in both schools was that the combined effect of reflectors and unshielded bulbs (i.e.,no lenses or diffusers) produced a very high contrast in the room. This may have resultedin students experiencing some degree of eye strain.

Table 4.15 records the percentage of smdents neverwearing glasses. Here it may be notedthat the greatest increase in the regular use of glasses by students occurred at Site 1from75.6 percent of students never wearing glasses at the beginning of the study to 65.5 percauat the end of the study. Blackwell (1985: 340-353) identifies two factors which may serveto explain the higher need for glasses. First, the spectral effectiveness of HPSVsuch that 7.5 times as much illuminance is required from them in order to match asoul= with a relatively uniform spectral power distribution. Second, the monochronniticHPSV lights (560 to 610 nm) may result in higher degrees of chromatic aberration andinappropriate focusing than the me balanced lights.

Conclusion. On the basis of an analysis of vision records for the childrenlocated in the different lighting environnwnts eyamined in this study, thenull hypothesis was supportedlight does not have an effect on changes invision. Nevertheless, it must be underscored that anwnalies in the rate ofchange in uncorrected vision at two of the sites point to the conclusion thatthe quality of lighting systems may be critical. Lighting systems shouldproduce comfonable light and 14:how high contrast levels.

Fluorescent Lamps as Ultraviolet Sources

The choice of Vita-Lites for use in this study has already been discussed. Nevertheless, thequestion renudns. Are there other fluorescent lamps that emit ultraviolet radiation? Theanswer is, Yes. All fluorescent lamps emit some ultraviolet radiation but this radiation isprobably little more than that which is common and characteristic of this type of lamp. Themanufacturers of Vita-Lites cb, however, claim to provide ultraviolet radiation in sonrwhatthe same ratio that ultraviolet radiation is present in daylight. A comparison of a randomselection of a number of readily available fluorescent lamps is presented in Table 4.16. Inthe UVA band which is of interest to this study, the Vita-Lites radiate almost 60 percentMOM energy than the nearest competitor. In the UVB band which is also of interest to thisstudy, but more of a concern from the point of view of safety, three of the lamp types emitmore energy than Vita-Lites.

Table 4.17 i/ data about the life or longevity of the phosphors used in Vita-Lites. Atthe end of t months of use, a sample of Vita-Lite lamps retained 61 percent of the UVAoutput and 79 percent of the visible light output. Stated another way, the UVA-emitting

hors appear to fall off at about the same rate as the light-emitting phosphors. Levelsof and UVC emitted by the lamps appears to fall off much more quackly with use.

Risk of Exposure to Supplemental Ultraviolet Light in Classrooms

Stepping outside into the sunlight means that we ate at the same time exposed to ultravioletlight and ultraviolet light is of three types. Two of these types are of significance to this

384 9

Table 4.15. A pre-test/post-test comparison of the percentage change in the number ofstudems never using glasses.

Pre-Test Post-Test Increment(1) ._._(29__t_LiE

Site 1 75.6 65.5 -10.1

Site 2 83.8 79.4 -4.4

Sites 3 and 5 68.0 66.4 -1.6

Site 4 58.6 55.4 -3.2

Table 4.16. Measured radiant energy from a sample of different fluorescent lamps. Allmeasuremans were made at a distance of 163±1 cm from the surface of thelamp. AU lamps were new and vien given a three to five minute warm-upbefore testing.

Lamp Type

Spectroradiometer Measurements"U V A UVB U V C

320-400 ma 280-320 um 200-280 umowan2 00=2 aW/cm2

Illumination2Photometric

=BeLes

Vita-Lite 40W 29.4 1.02 0.00 200

Duro-Test Canada, Inc.

140D Daylight 18.6 0.00 0.00 245

GE Canada

GTE F4OCW Cool White 17.7 1.70 0.00 320

Sylvania

GE F40-050 Chrome 50 13.7 1.70 0.00 210

GE Canada

GE F40WW 11.1 2.00 0.00 270

Warm WhiteGE Canada

Ultraviolet MY) Safety lanaiUV =Ration safety limits for an eight (8) hour period are as follows: (Phillips, 1983:4):

UVA (315-400 nm): 1000 11WitmzUVB + UVC (100-315 um): 0.1 pW/cm2

1111211111thlaglitaNni1. IL-791 Spectroradiometer System manufactured by International Light was used for UV measurements in

the narrow bands (UVA, UVI3. UVC). This NBS traceable calibrated system was used to conduct narrow-bind (5 um bandwidth) measurements of ultraviolet wave band from 200 to 400 milometers.

2. Model 40 Optometer manufactureed by United Detector Technology Inc. was used to measure theilluminance of the test lamps.

40 5

Table 4.17. A cotnparison of new lamps and lamps that have been in use for fifty (50)muds at study sites. All measurements wae made at a distance of170±0.5 cm from the suzface of the lamp. All lamps war given a thin tofive minute warm-up before testing.

Lamp Type

Spettmagioneter MeasnrementstUV A UVB UV C

320400 ma 280-320 nm 200-280 um

UW/cm2 uW/cm2 nW/cm2

Illumination2Photometxic

1518eLux

VLite 40W 30.0 0,78 0.05 201

Dwo-Test Canada, lnc.(Sample of 3 new lamps)

Vita-Lite 40W 18.3 0.24 0.01 158

Duro-Test Canada, Inc. (Sample of 7lamps used for 50 months)

Percent of output remaining inlamps after 50 months of use

0.61 0 .31 . 20 0.79

1. IL-791 Specooradiometer System manufactured by International Light was used for UV measurements in

the narrow bands (UVA, UVB. UVC). This NBS traceable calibrated system was used to conduct normw-

band (10 UM bandwidth) measurements of ultraviolet wave band from 200 to 400 manometers.

2. Model 40 Optometes manufactureed by United Detector Teclmology Inc. was used to measure the

illuminance of the test lamps.

study. Ultraviolet light in the region neatest to visible light (Le., 320-400 nanometers) isreferred to as tin UVA band. Next to the UVA band is the UVB band with wavelengthsfrom 280 to 320 nanommers. UVC represents the thitd band aml includes all wavelengthsshorter than 280 nanometers. UVC itradiance emitted from the lighting systems used inthis snkly was ml (see Table 3.1) and therefore of no consequence.

Sunlight reaching the earth's surface contains energy in both the UVA and UVB bands.Energy in sunlight in the UVC band (potentially the most dangerous form of ultravioletlight) is complmely filtered mit by the ammphms heft= it reaches the eanh and is of littleconsequence. Indeed, Thorington (1985: 36) presents evidence of an absolute cutdf at295 nanometers as a result of atmospheric absorption.

The amount a UVA and UVB energy contained in sunlight is punted in Table 4.1& Atnoon on a sunny day in June the UVA irradiance in Edmonton may reach 4,885 ttW/0122.At the same time the effective UVB irradiance may approach 0.44 AW/cm2. UVAexposures in October are apFaximately 60 percent of the Jim levels whrle the UVB levelsare down to the range of 40 yacent of the June levels. Exposure to ultraviolet light iscumulative and may be converted to annual exposures masured in joules per squarecentimeter (J/cm2). Table 4.18 includes maximum doses of ultraviolet lightfor an eight-hour period and these cunnaative exposures have been worked out based on200 days of exposure per year. Two hundred days instead of 365 was chosen sincestudents are in school only 200 days per year. Following this .) reduces the risk ofusing the students' entire annual UV quota in the classroom. I might conclude fromTable 4.18 that 5,760 J/cm2 of UVA and 0.6 J/cm2 of UVB are the maximumrecommended dosages for an eight hour daily exposure for a 200 day period.

Exposure levels encountered as a result of UV stwlements provided at Sites 3 and 5 arealso presented in Table 4.18. There it may be seen dat the cumulative classroom dosage ofUVA was 36 lieu and of UVB it was 0.04 J/cm2 The 66 6^ me to UVA amounts to 0.6percent of the maximum recomnumded exposure. On the of hand, the exposure to UVBamounted to about seven percent of the recommended maximum expcsure.

To put the UVB exposure at Sites 3 and 5 into perspective, it is important to make somecomparisons. First, the UVB emitted by the Vita-Late full spectrum lamps is greater thanthat emitted by other lighting systems involved in the study (see Table 3.1). Second it isalso important to compare the cumulative effect of five hours of akposure per day in the UVsupplemented classroom with the cumulative exposure that t be encountered by a 15mtnute akposure to the noontime sun in October or June in 6mon. The cumulativeclassroom UVB exposure levels were approximately 33 percent greater than a 15 minuteexposure to the noontime sun in October and about 50 percent of the exposure gainedduring a 15 minute exposure to the sun at noon in June.

Clearly, it can be seen from Table 4.18 that the UVA exposure levels in the schoolsreceiving UV lements were in the ?Inge of one percent a the maximum recommendeddosage. The exposure levels were about seven percent of the recommended exposurelevels.

Because of the inttoduction of UV radiation into the classrooms at Sites 3 and 5 it isimportant to understand some of the concerns that lie behind the recommended standardsfor UV exposures. One of the concerns is that skin cmcer rates in Alberta (and elsewhere)are on the increase. The rates in Alberta have more than tripled since 1971, with an averageincrease of 7.5 percent per year. This may be due to changes in ultraviolet levels as afunction of ozone layer depletion or it may represent the fact that the population spends

42

53

Table 4.18, Relative safety of UV supplements in classrooms,

Nature of UV Exposure

UV A UVB SEffectivel)

giVicm2

AnnualDose2i/cm2

AnnualDose2

W/cm2 / m2

Daylight Exposure LevelsEdmonton exposure 2925 0.14,October noondine

Edmonton exposure 4885 044June noondme

Recommended maximumeight.hour dosage

1000 5760 0 . 1 0.6

(200 days(yr)

Exposure ComparisonsSites 3 and 5 <1'3 <36 <0.01 <0.04Average exposures3

Ftfteen minute outdoor exposure 2925 526.5 0.18 0.03October noontime

Fifteen minute oustor exposure 4 4885 8793 0.44 0.08June noontime

1 The efficacy of uluaviolet Isthmian is not uniform as soss the spectrum. Maximum efficacy ways inthe 260 to 280 naaometer range. In order to pesmit comparisons,the measured UVB inadiance (shownin Table 3.1) has bent convened to effective irradiance in this table,

2 Exposure to ultraviolet radiation is cumulative. Theaverage annual dose tepresents ths cumulativeamount of exposure for a year expensed as joules per square centimeter (J/cm2). A wan of clergy equalsone joule per second. Cumulative exposure is calculated as follows

Expos= (J/crn2) = Effective UV irradiance in wausicm2 X exposure time in seconds

3 Inasmuch as students are in school for only 200 days per year, all calculations are based on 200 daysinstead of 365. Studems are exposed to ultraviolet relation for apiroximately five (5) hours per day for200 days per year.

4 Exposure equivalent to 15 minutes per day for 200 daysper year.

=ore time outdoors now than in 1971. It may also be that these are other factcas at work.

For . it may be die case that fewer of the pe°Ple walthig !,11f,tinbiffninflinie htlf:- to sunlight. As a consequence they may expericice more

those with "hardened" sidn when they do ventme outside. N - , one thing is

clearultraviolet radiation is considered to be carcinogenic and exposure elevates the risk

of skin cancer.

Though it is clear that ulunviolet radiation in the UVB band is beneficial in the production

of vitamin D in the skin, and vitamin D formed in the skin seems to be of a different %utility

than that which is taken orally, there remains the problem of establishing an oamount of UVB exposure. Until that optimum level is established (perhaps on the ,ofsone risk-benefit ratio), caution would su that any unnecessary exposure to UVB is

unwise. The debate is similar to that of and eggs versus cholesterol. How much is

good? How much is too much?

There is also a concern about the relative UVA:UVB imbalance when compared to natural

light which has a ratio in excess of 10,000:1. At Sites 3 and 5 the ratio was about 1000:1.

Tanning occurs in response to irradiation in the range of 290 to 340 nanometers (Patbak,

1985: 331). UVB irradiance (280 to 320 nm) stimulates vitamin D production, skinredda!ing and it initiates the tanning process. UVA irradiance (320 to 400 nm) darkens

sldn pigmentation and this tanning acts to block out excessive UV exposur!. In this study

the classrooms UVB levels may have been sufficiently high to aunually nigger a tanning

response. The unbalanced UV radiation emitted b the test lamps was too weak, however,

to effectively roduce tanning. Consequeutly, - s is could be exposed to UVB levels in

classrooms, which under natural light could trigger a oxmal tanning response, but because

of the relatively low levels of UVA the normal tanning response may not be completed. Ifthis is the process leading to pigmentation and UV blockage or Otection by the skin to

furtherUV irradiance, then it may be risky to expose students to 111 utic doses of UVB

without the availability of sufficient UVA energy to enable the sldn to elop pignvntation

as a control for excessive UV exposure.

A further word of caution must be introduced in this section. Three of the fluorescent

lamps tested in Table 4.16 had highcr levels of UVB and lower levels of UVA than the

Vita-Lites used in this study. For these lamps, the UVA:UVB balance is inferior to Vita-

Lites and for that reason care and cauticai should be exercised, especially if these lamps are

to be used in fixtures =lifted to release UV radiation.

I II II 111

Summary

This study sin out to test the null hypothesisdifferent light types (full spectrumfluorescent, full spectrum fluorescent with UV enhancement, cool-white fluorescent, and

high pressure sodium vapor) have no differential effects on Division II students' dental

histories, growth and development histories, vision histories, scholastic achievement

histories, or attendance histories when compared over a two-year period.

The significant findings of the study are summarized in tabular form in Table 4.19.

Without placing a value judgement on the findings (i.e., good or bad) two symbols are

used to &saibe the findings in Table 4.19a negative sign () is used to indicate the

lowest or smallest measure while a positive sign (+) is used to indicate the highest or

greatest mess= This non-evaluative view must be kept in mind, especially when viewing

the findings with respect to the onset of maiarche and gains in body fat. Sites 3 and5 (full

spectrum with UV enhancement) and Site 2 (full spectrum) appear to contain the preferred

lighting systems inasmuch as they are significantly better than other sites on the greatest

5544

Table 4.19. A summary of significant findings of the study to with a desaiptionof lighting system charactezistics. Significant It (pc0.05) werefound to occur between negatives () and positives (+).

Factor ExaminedSite 1 Site 4 Site 2 Sites 3/5HPSV CW FS FS + UV

Lighting System CharacteristicsDominant color characteristic of light Goklen Yellow- Daylight DaylightSO= (yellow- gmen (bluish) (bluish +

aninge) UV)

Color temperature (K)1 2100 4250 5500 5500(Daylightsun and sky 5000)

Color renderin index' 21 62 91 91(100 natural daylight)

Non-Visual Effects on StudentsRedaztions in the dcN clopunent of dentalcaries.

na* _ +

Auendance ratio. + +

OnSet of menarche + +

Gains in height. + +

Gains in weight. +

Gains in body fat. + + +

Academic achievement** + + +

Site 4 was dropped from the dental component of the study when it was found that many studentshad received applications of fissure sealants.

Site 2 had significantly greater achievement gains than all other siteswhile Sites 3 and 5 and Site4 were only bent: than Site 1.

1 Thsington (1985: 50).

45

5 6

number of measures. At the seam time it must be noted that Site 1 (high pressure sodiumvapcx) rated significantly power on most measures.

From Table 4.19 it may be concluded that the null hypothesis (set forth above) wasrejected. Different light types do have differential effects on Division II students' dentalhistories, growth and development histories, vision histories, scholastic achievementhistories, or attendance histories when examined over a two-year period.

The findings of this study point to the clear conclusion that light has important non-visualeffects on people. Just how these effects occur is still uncertain.

When it comes to the findings in Table 4.19, it seems clear that IN supplements mayaccount for differences in the rates at which dental caries developed and this may be linked

to the stimulation of vitamin D 'on in the sitin as a result of UVB irradiation. Anumber of the other significant . .4 may be more related to color or the visible lightspectra than anything else. The onset of menarche seems to fit this case. The higher thanexpected incidence occurred at schools with full spectrum light (i.e., enhanced blue). The

lower than expected incidence occurred at the site with the yellow-oranp (nearmonochromatic) light. Indeed, gains in height and achievement awl attendance ratios all fit

into this same patternstudents under conventional or blue-enhanced lighting scored thelargest gains and those students under HPSV lighting scored lowest.

©Rtsgnscr

Conclusions and RecommendationsConclusions

Every study tries to answer some questions and to provide some new information. Inpursuing these objectives, it is often the case that the new information identifies still otherareas which need to be studied. Such is the case with this study.

What Do We Now Know About Light and It's Effects on Students?

On the basis of this current study, and other research 1-1 in the literature, we knowthat trace tumults of UV in the classroom have the - of reducing the incidence ofdental caries in children in Grades 5 and 6. We also suspect that the color of visible light inclassrooms may have an effect on the growth and development rates of children.

What Don't We Now Know About Light and It's Effects on Students?

We do not know enough about the risks associated with different lighting systems. Mostimportantly, we do not know where the boundary lies between risks and benefits ofultraviolet LOt.

Putting the Study Findings Into Perspective

Lighting systems are nwst often designed with efficiency in mindthe objective being toobtain the highest possible lumens/watt ratio. Seemingly little attention is paid to any non-visual effects lights may have on the occupants using the lighting systems. Clearly, thisstudy points to the single conclusion thatno matter how eftientlighting systems arenot neutral with respect to their affects on people. Indeed, it appears to be the case thatthere are a number of non-visual effects associated with lighting systems. This study hasidentified a number of such effectsdifferences in the rate of dental caries development,differences in rates of attendance, differences in the age of the onset of menarche,differences in height, weight, and body fat gains, and differences in scholasticachievement. One might conclude from these findings that natural light is u. to thedevelopment and well-being of people and to imprison people in spaces only withartificial lights designed solely for efficiency amounts to a clear case of daylight robtay.

Wittman ( 1985: xi) vay articulately describes the need for clear policy-making guidelinesand better understanding of light and its effects on people by first posing a question,Should limits be placed, based on health considerations, on the artificially illuminatedlighting environments under which people may live and work?

The question is important for all of us, but &specially so for people whocannot readily walk outside for some fresh air and sunshine--iike schoolchildren in very cold climates, or residents of nursing homes, hospitals,submarines, or spacecrqft. How much evidence, and based on whichspectfk tests, ought to be required before a given light source Ls adjudgedsafe, when that source may be used to proli& people with all of their lightfor many hours at a time, on many days? For years now companies thatmake artificial foods have labored under the burden of fortifying theirproducts with the vitamins or minerals that people would otherwise have

58

I. 1.. ! 1 1-'1 f. t lift,L! 211 I g I li!..1_.' I s s 1 !I q_..!_. 1 '

.4 I 11 ' I ....1.1.ii J 1 I I ....,1 1, . f I f f I

_!. ILL!: # I .44 ' I I 1.._ I t

I A _ALAI., I I I I I , I I '' I a. I t 1 :4_0:2,1 . ..4 I 4 f f

1 IL ,I II 1 I I .." l I 11.11 1 I s I I I

I . 4 i I I f 1 I /way / f VI,.

s' I I I a r p t 411 eV!

51 1 I' i ...., # at e :#.1 __I !.# , I 1 ' .41 I I I I , I

I fa

I . 5 ' .4 1 I Ifir I ,...: I III 1,11 a ' 6 111 1 6.46,0 I f el 1 ,..1,.. II 1 I I

VAIO .1 ''...L., a ....0 (6 a .3 a 61 .5 I 01 I A...... I II 0,6 ' I ii 6.101_, : ha I..I SI 1 Oil-

.: .4- -1'.11,. I '_.....0, 1,5 .5 .. : a ..., a I I I ' i, ;1 a. a' 11 I ....,0 t a 6116 tii ..1 11 I ....

111 'II 6

1

1 1 (11 6 6 11 11 11

s . A is a .110,./a, Is 00 S ..at.45 at :

1 11 1 0_4 4..40 11 1 , 11 .4 1 . I . 5 I 6 I4I I .1 11,, 1 4 ..o':

1 .1,4 1 t: . .4 1 ' "tt" 1.4 '6 It : 11161.4 a 1 11 1

1 .4 6 t e , I 1 fr: 1 0 Itt 1 'Int ` : : t !. I g I s.e..._, ,',1 e i a -(t 1 .4 : t a I $ ..4

t e 1 1 ..... :' 1 1 _146 1 6 e,_ 14.1 ...11 II It fills 161 ..4 r4112 6 !. f_

I I 010 1t.ja a a I I I /$.1',P.1 I. 1 .II 1 11, '4 I 111 1 ...... .._ L I

I 6 11 II a t 6 .41 0 $ n1 6 f 1 1 I 0 : ,5 .41 6 I '. 1 " 4 _ t4 6 I_

1 a 1 1 t 6 II If ' 0.1 : I I .11 jet 41 6/1 '1 II a Ve 11 . 6 I St' Lt.* a 6 0 : : IC

"A 10 et, I 0 I '40

0 eat el r6 .111.41

1. ' 11 . 1 1 : .. I I ts la 6 Olt'. I 6 e ' ...4 I 1 1 11 :6 II. 6. 111 a 61 Vall ' 4 1 1 . 1 /I : 1 :, 11 I .1 It 1: 1 1111. 1..,011111 et

Vs I r. 11 1 11_1: ,11 51 : e I 1 1 "16 I_LLI 1 r. 1 I : 11 II 0 064111114 ', I 1 11,1 I : 1 ft:

11 . 1 a - - I I I ...., I .; I : 'II IS . ' Oast i .4 .' : a ta

0 11 : ..11S I, 10.4 .6 11 0 1 : 11) 1 , 1 II A. III . a

1 .., . 11111 s : ,,1t646 11 II 1 '4.. ' I .11 ft 6 If 1.6 I 11 0 a .O I/ I .... 11.., a lk 111 at .

I 0 II I ; 164 a : 1 : 1 a 11 t I I 64.7.443._ 1 0 1

6.6 II as ...it 4 .11 fi . at III -4 : 1 $11 I 0 . 1k 11 0 I i VII 5 ' .4 10 : 1 .....

. a 1_1 II16,1016: a la I I +.1 al I fit 1 111 ..0 6 5 (...11 55 1 . I i I sin. 1 I 5 .6 16 I 4 .1 04..1 V 0 11, .10,6 11111 .1 6,-6 11, II II

1 .1 11. 1 6 a 1 I 1 ' 1 ' 1 : a a a a

.0 1111 tat.- 1 r II I 1 .5 . 1 11 I a Oa .0) , 1 O.L',6

e.01:111!1_ r . $ : 1 O pill I II :Is II

11 .4 .11 vt 1 v.i 1, . I 1 1 I , 11.11 e: 1

1 .1 0 a b' I..e

0 , " l a -.11.4 .. a a r ' a i w at ta

a 1 -- ...; -, t l_a_tbs1 Li as 1 a ' 6 *.11 A t 1 t i a a ' 11 a 1 a -4 la a as 5..

a 11 ' II 1 svt 1 .-1 0,.4 .1 O I I ' 1 11 I ie .4 IA.....1 A 9, a t 1 I

: a '4 161 a a 1 WI : 11 a se 6, f,v I I. 1 01[ 6 1 0 et Llakla P. _._11 I

It 1,11 ' I 1 I 1111 i 0 1 . II .-, 1 1 14 ` 1 I 4 -0 4 11 11/..1 II I *1 114

.11 4 1 . t11 11r, `m4-4 1 11 111.11 111 51+4 1 4 . I ..1..4 A I I v lie 'I .

11......11 1 t 1 II 1 i -11, - 11,111111 .+1 4 ..-11 111 4 /1 III t,.. ei. L a i 1 I '11 1 SI 1

I..r.q_,.. I I I ' ktiet I tt ' 4.44 *NA ..., 'JI 1 1 I ILIZ 0.1 44 II ',.. . 11.... . II II 1.,..±1_1t,01 1 114 .111 4 III .41,10 4 It .. .14 ' 1 Il'i I I i t' 1

1 I I 1.6111111 41111111 -.

4 '111 0 51 4,1 144 It 1 4 I 121.1).6 If ' I i .41.0.... I 11)

i -4 !...tt-1, It :,!._,-.11 i `' III ' 1 . 11 1 .1 : "-4 t i 14 WI 11 1 , 11

4 .., 1 11 : , 1 0 Silt I ....1 it , 11 :a1 I :..! t *1 1 4 * , 4. 4.4,1,4,

I I 4 1 611;106 .111 I I 1 0441.21,A,..4 ...t I 1.4 It.: 5 .4. *11 1 40

j : I ' A I I, 1 WI 4 i 0 I .: 4 1 1 1 0 i . " 9 1 A 16 I 4 I 4 4 I 5 1 5 5 .:11111111 .4 111 ..11

I 4 ' ' 4 0 i 6...110, 5 1,114111i. WI 14 IIII 4,A 1 I 1 4"( 11 11

1.4 I 6 III I 1.61,1 A I It 1 0 6 I I 1 I 11111..4 1 li : 0

.5 1 .411 1 6 41. III I ,El lit wi ..5 1 5 5, 5 5 I : : ik,.1 9....* $ I ...It

I I 1 ' 4 .541 .0 41, I I I S .4 I . $ I I I

6. vt .1 1 0 1 II I ' .0. t) I I II I .11 ' ket I 4 I 5 II4, 5 !At. I I 5 5 11 : 4 : 4 I 1 i I ' i 51 ,Ik0 V I 1.1

ft II: 116 6461`10.1IIII .1 I I 11.61'15

51 444 5 .1 o o- III I P I Is 4 I 1,14 .11

I ,011 . I 1...11,111.0.1 , I 4 . SI ' ` 1 4 41 1 . IA 41 4,..4 : p: 44, - 1.1 i 516 4 5

I .4 t I I . I II 1 0'6411. 1 4 I r ,I .4, 1 : ...-,1 0 II, 1 vsllni 1, S. II .41:1

6.: III $l 0 I 1.64 k '-leti II,* 5 A. 5 I III I 10 II I .,..1 .1.4 e ate . ,....;4 4_

.4 A L . - 1 A 1 ' 5 I 5 : . 4 : 5 5 1 ..., 1 I ' 11 I I 11.04_i : 8 11,0 I

0 5 I I : 1 1 1 0 I . '1 II I II 1 4 I el .5 1 Ity4

wi I I . 0 . I ltIkt 4, .0: : I 1 . 4 ' 1 .4 I 11 I 4 4 .41 10 lio 0 ' 61112. 1 4,,4

I 1 6 . . 4 5 5,4 1 1.4 .0.11 I "L

t :I III4: i 41 I ' II III I IS 514 .1 1 I * :I

_._". II I 1 1 5 16 1-11111 10 $ 11 ....10 1.4 1 11 I I 'i1e 4_, 4 I I 4 44 I1 ,1,..11 4, i ' 1 0 I 6 44 I.., 61 II ' 0 I.'. III 5 111 1 i I It I: 1r4.11. I '

III ` 1 4 4.4 i y...M14 1 :I 1 I 1: III 1 II 1 11 1 I.11 le 5 I ... 5) If 4 ' 0

II t 4 ..1 1 4 1/1 1 ,I 0 11 1/ 1 S .44 I II . I I ' 5 5 t I '1 :AI I I i 1

.., 1 ' I I 1 5 .5 I ' :- :.. 3 1 : I 11 : 1 0 5 1 ar: 11 : II 5 1 I 1. 5

41 III I , /I If ,1 3" 1 I I II 1 .4 I $5111' 1 el ....:$ 5 to 5 ojjgo,, II - 4 II : I - . 1 . n II 6 11 I : ' ... `.11i

1 I I ....., '1 I

.5 I .101 : I I ..4111..j1 1141 4. 4

a 'I I a i . .8 .1i1 ii 4510 1.11 $ ' I .4 11 I II -4 I.11 1

4 4 * ; -.al 44 f 01 1..4 1 'II ' 4 11.. 11..4 I ' .1.11 ' 6 I a 4, I : I o I , 11 4,1

.4 4 1 1, 1 4 1 1/.0.410,1 01 4 : : 1 1 1 4 11 1 1 II ...4 11 4 4 4 41 I 4 /1

.11 I 14 1,0 tll 11 1411 II 14 1 C , 1 II 0 11 II. 4 Ili 41 I L I 1 II. I 1 ,:1 1* 0 VS 1 1

/11,11._1611.1 I I 0 .0 11.0,1:111k (0 'Ill/ %II '14.1 I 1 4 . 14 : 4. -11) 1 1 4,4 'II .4 1 41,1411 (4 1 . 1 I 1 I 4 el .4 4.L.. 4 . e fit: 4jt0 ,t't ::,....41 .4 4t..114.`/.11.1....,, IS

I . 5 .1 41 III I'' : I 1 * I I 1 i 4. I 1 . 4 -ie l: 1 1 .1 1 1 11*:

4115,5:44 I : 4 1 *IiI 01 0 4 1,011111 0 III 11 S I II; 0 O 4 .0I 1 II I I 0 I

-4 6 61 .4 4111 i 11 4,-4-, . 10 I A Oki ' .1 ' II 0 ,II VII ..1 411 111111 3 0 4 ' 1 IS.-

1 1 1 111 1 41 0 I 5 . 111 ' 1 4 1

6. What are the implications of this research on the workplace? On health facilities? Onfacilities for the aged? On osteoporosis?

This study has focused on children from ten to 12 yearsof age. Accordingly, the findingscannot be generalized beyond this age group without some deg= of risk. Nevertheless,the research findings maned in this study sugpst that similar effects should be looked forin other segments of the population. Obvious targets for future research should be in, (a)offices where - le spend most of their daylight hours, (b) in health facilities, and (c) infacilities for aged. It is a well-established that the aged lose their ability to absorbcalcium with the result that they tend to suffer fzom osteoporosis and increased boneporosity and fragility. This segment of the population may especially benefit from lighttherapy.

The second set of questions pertains to ways of using the findinp of this study.

1. How should tiv schools be modified to capitalize on findings of completed research?

Clearly, the selection of classroom lighting and colors should be based on something monthan economics and energy conservation. The big question asks, What can facilityplanners do, based on the available research, to improve learning environments? Theanswer is several things.

For people who are outdoors for a significant part of each day, the quality of indoorlightinf to which they are exposed may be of little consequence. Their needs for naturallight stimulation may be adequately met. But for people who spend almost all of their timeindoors, and with outdoor exposure limited to morning and evening light, there may be aneed for artificial lighting that is supplemented with light stimulation in the spectnnn areasof energy deficiencyespecially ultraviolet in the 290 to 400 31113 range (UVA and UVB).

Using windows to obtain the required ultraviolet light supplements is of marginal effectbecause some ultraviolet light, especially UVB, the most significant missing component inartificial light, is almost completely filtered out by ordinary window glass. Allen (1971:33) reports that glass attenuates UV transmission below 325-350 nanometen while acrylicand styrene have sharp cutoffs in the range of 375-400 nanometers. Moreover, natural UVwithin classrooms would be available only in full sunlight and near a window. Much ofany UV passing through the window glass and diffused throuut the classroom wouldbe lost due to absorption by surfaces in the classroom. (*/y materials absorb UVradiaiion [see Table 2.2]). To create a truly balanced lighting in classrooms, pahapscontrolled amounts of ultraviolet radiation should be introduced into classrooms as a panofthe overall classroom lighting system.

Color may also be used to advantage. A rule of thumb suggests that cool colors fosterrelaxation and warm colors foster stimulation. With this in mind, planners might ny todetermine the sorts of activities to take lace in the learning spaces and to plan colorsacconlingly. Active spaces like _ ums might benefit from extensive use of warmcolccs. Libraries might be suits candidates for the cool colors. The same may be truefor washrooms, showers, and corridors. Spaces for over-active children or cldWen withbehavioral 1Problems mi be decorated with cool colors. Students who are slow awlunmotivated might -1 from warm colors.

Finally, on the basis of the sparse research into the effects of light from high pressuresodium vapor lamps on people, it is probably unwise to use these lamps in regularclassrooms. Their use should be limited to gymnasiums, corridors, and other spaces usedonly occasionally by students.

50 61

2. In what ways can mice amounts of UV be introduced into classtooms?

If it was decided that UV radiation should be introduced into classrooms, there are seventlways by which =trolled amounts might be intmduced. One way might be to providebanks of sunlamps in some romns or areas frequented by all Waing cxxmpantsareassuch as lounges, cafeterias, showers, washrooms, and gyms. A second way might be toreplace existing classroom lights with UV emitting full-spectrum fluorescent lamps.However, without rrilectors and special UV transmitting diffusers CC WISCS the UV energyis absorbed by the fixture and thus serves no useful A third way might be toexplore the use of ultraviolet fluorescait lamps (black _4, ) mounted in special fixtures inthe classrooms. If these lamps emit ultraviolet in the appropriate band and at &shed levels,they could be used as an alternative to the use of the more costly full spectrum lamps andthe nxxlified fixtures that they requite when used as ultraviolet sources.

An alternative to changing lighting systems might be to supplement diets with suitabledoses of vitamin D (for example, by &initing two to four glasses of fortified milk daily).Wunman (1975: 474) does, however, question whether dietary sources of vitamin D arebiologically as effective as the vitamin D fonmd in the skin and the smdies reported byAlberta Education wad to strengthen Wurtman's concerns. (Most of those studentsinvolved in the Alberta studies received reasonable levels of Vitamin D through theirdietsespecially through milk which is fortified with Vitamin D.) A second alternativemight be to provide some outdoor activity between roughly 10:00 AM and 200 PM(especially during the winter months).

It is important to temember that unless reflectors and UV transmitting lenses are installed in=junction with installation of UV emitting full-spectrum fluorescent lamps, the onlyadvantage to be obtained frmn full spectrum lighting is impnwed colic rendition. In thisregard, full spoon= lamps may be used to pmvide high quality lighting in paint shws,artstudios, and other places where correct color rendering is important.

3. Who are the principal benefactors of the beneficial effects of improved lightingsystems? Who should pay?

One of the clear benefits that could derive from applications of this research into thelearning environments children occupy could tx. reductions in the development of dentalcaties. The benefits of this action would accrue to parents in the form of reduced dentalbills.

The second clear benefit derives from the fact that under improved lighting conditionsstudents seem to learn better (i.e., demonstrate better academic achievement) and theirattendance improvesstudents are occupying the spaces provided for them (i.e., improvedattendance) and they are learning more. The combined effects of these two benefits clearlyaccrues to the educational institution.

Recommendations

This study was undertaken for the purpose of exploring the effects gm elemen schoolchildren (Grades 4, 5 and 6) of diffetent lighting systems. The findin of the y affirmthe findings of an earlier study (Wohlfarth, 1986; Hargreaves & . , 1989) andpoint to the conclusion that lightfar from being neutral with respect to effects onchildrenhas significant non-visual effects on children. The study pAnts to differences indevelopment of dental caries, the age at the onset of menarche, gains in height, gains in

weight, gains in body fat, gains inghti

achievement, and improveurnts in attendance as beingrelated to classroom ling types.

Several reconunendadons flow logically from this study.

1. When it cmnes to lighting sources, this study suggests that there are cleardiffernicesin quality. Full - - lamps (Le., Vita-Lites) rimmed in ultraviolet transmittingfixtures provide I with demonstrable beneficial effects. Full spectrum lamps inconventional fixtures also appear to contribute to differences in growth anddevelopment patterns. Cool-white fluorescent lamps mounted in conventional fixtures

to be neutral with respect to many of these non-visual effects High pressurevapor lamps appear to have negative effects onchill:ben.

Inasmuch as cool-white lamps or full spectrum lam* mountaxl in conventional fixturesemit insignificant amounts of UV radiation and because have no identifiabledrawbacks, they should continue to be used.

High pressme sodium lamps should not be used in classrooms until further testsprove that they have no -4i ects on children or the way they behave.

2. Ultraviolet levels within the UVB band should be checked in all schools equipped withfluorescent lamps (any brand) mounted in lighting fixtures that have UV-transminingdiffusers. Included in this category are the rather common egg-crate diffusers. Whenmounted in these fixtures, any fluorescent lamp can emit some ultraviolet radiation.

3. Schools and school districts should be cautious when it comes to changing lightingsystems to match the UV-emitting systems used in this study. There needs to be abetter understanding of the relative risks and benefits of UV in the classroom.

4. When the more expensive full spectrum fluorescent lamps are to be used as a source ofcontrolled ultraviolet light, care must be taken to ensure that the chosen lamps emitreasonable levels of ultraviolet radiation at desk top heights (probably in the nage of 8to 10 iiW/cm2 as a minimum). Reflectors should be selected which have noultraviolet-inhibiting coatings on the front surface. Laboratory tests can be used toverify that reflect= Qperate in the ultraviolet range. Fmally, lenses should be capableof transmitting ultraviolet radiation. Again, laboratory tests can be used to determinethe ultraviolet transmission efficiency.

5. When fluorescent lighting fixtures are modified with reflectors and ultraviolettransmitting lenses for the purpose of obtaining an output of ultraviolet radiation, caremust be taken to ensuie that the ovetall quality of the lighting system is not degraded.The addition of reflectors, in particular, has the effect of increasing the ccaurast ratio ofthe lighting system and it is suspected that this may contribute to significant eyefatigue.

6. Three of the fluorescent lamps tested in Table 4.16 had higher levels of UVB andlower levels of UVA than the Vita-Lites used in this study. For these lamps, theUVkUVB balance is infetior to Vita-Lites (and much worse than natural light) and forthat reason these lamps should not be used in classroom ting fixtures which havebeen modified to release UV radiationat least not until "tional studies have beencarried out.

7. Further research should be undertaken into the non-visual effects of light in at leastduce areas.

Research should be undertaken in order to explore optimum lighting systems (i.e.,person lighting and not just task 14hting) and opthnum levels of ultraviolet light to beintroduced into classrooms. The ideal 111. t be to conduct a cow:oiled study in anurban district where half the schools co be used as experimental sites and half ascontrol sites. The ability to control for a number of factors (especially socio-economicfames) could be easier than in this study and ifs 1981-1985 predecessor.

Research should be undertaken to determine the most efficient and cost-effective waysof intmducing controlled amounts of ultraviolet radiation into classrooms.

Research shcold be undertaken in order to explore the 4eneralizability of the fmdingsof this study to other t ti'lulations (pre-school, the primary grades, and junior andsenior high school) and I geographic locations and climates.

8. In the field of lighting and lighting effects (which is rapidly expulding) there is a needfor a clearinghouse of information and ideas for implementation. A clearinghouse ofresearch into the non-visual effects of light and color and implications for schoolfacilities should be established. Funding of future studies should be integrated withother facility-related research and synthesized to yield clear school facility designguidelines.

BibliographyAllen, C J. (1971, January). Color temperatures. Illuminating Engineering, 31-36.

Aston, S. M. & Bellchambers, IL E (1969). Mumination, colour rendering and visualclarity. lighting Research and Technology. (1), 259-261.

Bickford, E W. (1981). Nonvisual effects of radiant energy. 1ES Lighting Handbook1981 Awlications Volume. 19.1-19.40.

Blackwell, H. R. (1985). Effects of light source spectral distribution upon visualfunctions. In R. J. Wurtman, M. J. Baum & J. T. Potts, Jr. (eds), The MedicalEffects of Light (Annals of the New York Academy of Sciences, Volum 453).New Yadc The New York Academy of Sciences.

Corth, R. (1984, January 19). What is "natural" li t? A paper presented at theConference on the Photobiological Health Effects of Artificial Li t Sourcesconducted by the Puget Sound Section of the Illuminating En Society,Seattle, Washington.

Davies, D. M. (1985). Calcium metabolism in healthy men deprived of sunlight In R. J.Wurtman, M. J. Baum & J. T. Potts, Jr. (eds), The Medical Effects of Light(Annals of the New York Academy of Sciences, Volume 453). New York: TheNew York Academy of Sciences.

Downing, D. (1988). Day Light Robbery. London: Arrow Books Limited.

East, B. R. (1939, July). Mean annual hours of sunshine and the incidence of dentalcaries. American Journal of Public Health and The Nation's Health, 29, 777-780.

Faily, A., Bursor, D. E. & Musenieche, R. A. (1979, September-October). The impact ofcolor and lighting in schools. CEFP Journal, 17 (5), 16-16.

Ferguson, G. A. (1971). Statistical Analysis in Psychology & Education (Third Edition).Toronto: McCnaw-Hill Book Company.

Hargreaves, J. A. & Thompson, G. W. (1989). Ultraviolet light and dental caries inchildren. Caries Research, 23, 389-392.

Helveston, E. M., Weber, J. C., Miller, C. O., Robertson, IC, Hohberger, G., Estes, R.,Ellis, F. D., Pick, N., & Helveston, B. H. (1985, March). Visual function andacademic performance. American Journal of Ophthalmology, 99 (3), 346-355.

Himmelfarb, P., Scott, A., & Thayer, P. S. (1970). Bacterial activity of a broad-spectrumillumination source. Applied Microbiology, 1013-1014.

Hodr, R. (1971). Phototherapy of hyperbilirubinemia in premature infants.Ceskoclovenska Pediatric, 26.

f;5

Ho lick, M. F. (1985). The photobiology of vitamin D and its consequences for humans.In R. I. Wunman, M. J. Baum & L T. Potts, Jr. (eds), The Medical gliects ofLight (Annals of the New York Academy ofSciences Volume 453). New York:The New Ymit Academy of Sciences, 1-13.

Holick, M. F., McNeill, S. C., MacLaufhlin, J. A., Holick, S. A., Clark, M. B., &Potts, J. T. Jr. (1971). Physiologic implications of the formation of previtamin1) 3 in skin. Tranuactions of the Association of American Physicians, 92.

Hughes, P. C. (1981, Maivh-April). School lighting fix the total person: a psychobiolgicalapproach. CEFP Journal, 19 (2), 47.

Inpaham, L. (1983). Electromagnetic Radiation and Student Off-Task Behavior.Edmonton, Alberta: Planning Services Branch, Alberta Education.

Kerlinger, F. N. (1964). Foundations of Behavioral Research. Toronto: Holt, Rinehartand Winston, Inc.

Lamola, A. A. (1985). A history of organizations interested in biological effects of light.In R. J. Wunman, L Baum & J. T. Potts, Jr. (eds), The Medical Effects ofLight (Annals of the New York Academy of Sciences, Volume 453). New York:The New York Academy of Sciences, 121-122.

Loomis, W. F. (1970, December). Rickets. Scientific American, 223 (6), 77-91.

Lucey, J. F. (1972, November). Neonatal jaundice and phototherapy. Pediatric Clinics ofNorth America, 19 (4), 1-7.

Maas, J. B., Jayson, J. K. & Kleiber, D. A. (1974). 'Quality" of light is importantnotjust quantity. American School and University, 46 (12), 31.

Neer, R. M. (1971, August). A paper nted at the National Technical Conference ofthe Illuminating Engineenng ty.

Neer. R. M. (1975). The evolutionary sigtificance of vitamin D, skin pigment, andultraviolet light American Journal of Physical Anthropology, 43 (3), 409-416.

Neer, R. M., Davis, T. R. A., Walcott, A., Koski, S., Schepis, P., Taylor, I.,Thorington, L & Wurtman, R. J. (1971, January 22). Stimulation by artificiallighting of calcium absorption in elderly human subjects. Natter, 229.

Ozaki, Y. & Wurtman, R. J. (1979). Spectral power distribution of light sources affectspowth and development of rats. Photochemistry and Photobiology, 29, 339-341.

Painter, M. (1976). fluorescent lights and hyperactivity in children: an experiment.Academic Thempy, 12 (2), 181-184.

Pathak, M. A. (1985). Activation of the melanocyte system by ultraviolet radiation and celltransformation. In R. J. Wurtman, M. J. Baum & 3. T. Potts, Jr. (eds), TheMedical areas of Light (Annals of the New York Academy of Sciences, Volume453). New York: The New York Academy of Sciences.

56 f; 6

Phillips, B. G. (1983, May 3-5). Ultraviolet radiatim and fluorescent lighting. A paperpresented at the 4th Annual Conference of the Canadian Radiation ProtectionAssociation, Torceto, Ontario.

Sharon, L M., Feller, R. P., & Burney, S. W. (1971). The effects of lights of diffaentspectra on caries incidence in the golden hamster. Archives of Oral Biology, 15(12), 1427-1431.

Sydoriak, D. E (1984). An eiperiment to determine the effects of light and color In thelearning environment. Unpublished doctoral dissertation, University of Attansas,Little Rock, Arkansas.

Thorington, L (1985). Spectral, irradiance, and temporal aspects of natural and artificiallight In R. J. Wunman, M. J. Baum & J. T. Potts, Jr. (eds), The Medicalffects of Light (Annals qf the New York Academy of Sciences, Volume 453).New York: 11= New York Academy of Scion:es.

Thorington, L, Cunningham, L & Parascondola, L (1971, April). Visual and biologicaspects of an artificial sunlight illuminant Illuminating Engineering, 240-250.

Thorington, L., Parascondola, L. & Cunningham, L (1971, (ctober). Visual and biologicaspects of an artificial sunlight illuminant. Journal of the Illuminating EngineeringSociety, 3341.

Volkova, N. V. (1967). Experience in the use of erythemic ultraviolet radiation in thegeneral lighting system of a machine shop. Translated by Duzo-Test Electric fromoriginal Russian in Gigiena in Sanitariia, 3Z 109-111.

Wohlfarth, H. (1986). Color and Light Effects on Students' Achievement, Behavior andPhysiology. Edmonton, Alberta: Planning Services Branch, Albena Education.

Wurtman, R. J. & Neer, R. M. (1970, February 12). Good light and bad. The NewEngland Journal of Medicine, 282 (7).

Wurtman, R. J. (1968, September 9-12). Biological implications of artificial illumination.A paper presented at the National Technical Conference of the MuminatingEngineering Society, Phoenix, Arizona.

Wurtman, R. J. (1969, January). The pineal and endocrine function. Hospital Practice, 4(1), 32-37.

Wurtman, R. J. (1975). The effects of light on man and other mammals. Annual Review ofPhysiology, Volume 37.

Wurtinan, R. J. (1985). Introductory remarks. In R. J. Wurtman, M. J. Baum & 3. T.Potts, Jr. (eds), The Medical Effects of Light (Annals of the New York Academyof Sciences, Volume 453). New York: The New York Academy of Sciences.

Wurtman, R. J. & Weisel, J. (1969, December). Environmental lighting andneuroendocrine function: relationship between spectrum of light source andgonadal growth. Endocrinology, 85 (6), 1218-1221.

Zanikova, M. A. & Krivitskaya, E. I. (1966, April). Effect of irradiation by ultravioleterythema I.. on the worldnf ability of school children. Translated by Duro-Test Electric 'ss original Russian in Gigiena Sankt:Ma, 31, 41-44.

58